CN116107815A - Internet data backup method, device and server - Google Patents

Internet data backup method, device and server Download PDF

Info

Publication number
CN116107815A
CN116107815A CN202310364120.9A CN202310364120A CN116107815A CN 116107815 A CN116107815 A CN 116107815A CN 202310364120 A CN202310364120 A CN 202310364120A CN 116107815 A CN116107815 A CN 116107815A
Authority
CN
China
Prior art keywords
priority
data
carrier
level
father
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202310364120.9A
Other languages
Chinese (zh)
Other versions
CN116107815B (en
Inventor
罗颖川
卫彩霞
陈林招
彭水平
梁邦领
幸杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Microgrid Union Technology Chengdu Co ltd
Original Assignee
Microgrid Union Technology Chengdu Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Microgrid Union Technology Chengdu Co ltd filed Critical Microgrid Union Technology Chengdu Co ltd
Priority to CN202310364120.9A priority Critical patent/CN116107815B/en
Publication of CN116107815A publication Critical patent/CN116107815A/en
Application granted granted Critical
Publication of CN116107815B publication Critical patent/CN116107815B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1446—Point-in-time backing up or restoration of persistent data
    • G06F11/1458—Management of the backup or restore process
    • G06F11/1464—Management of the backup or restore process for networked environments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1446—Point-in-time backing up or restoration of persistent data
    • G06F11/1448—Management of the data involved in backup or backup restore
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Telephonic Communication Services (AREA)

Abstract

According to the Internet data backup method, device and server, the priority of the father level and the son level of the Internet data to be backed up is divided, and the data with high priority can be backed up preferentially by carrying out data backup according to the priority, so that loss caused by emergency is reduced. In addition, in the process of dividing the data priority, the internet data backup network is adopted to conduct data priority identification and determination, so that the accuracy and the efficiency of data priority determination are improved, and the rationality and the efficiency of data backup are improved.

Description

Internet data backup method, device and server
Technical Field
The present invention relates to the field of data processing, and in particular, to a method, an apparatus, and a server for backing up internet data.
Background
The data backup is the basis of data disaster recovery, and in order to meet the requirements of data safety, traceability and the like, the internet data needs to be backed up regularly. In the prior art, for internet data backup, the backup is usually directly performed according to the sequence of the data, and under the condition of large data volume, the backup time is usually long, and if an emergency occurs in the backup process, the backup middle section is caused, so that the data may be lost. In addition, for the backup scene with high real-time requirement, high requirement is put forward for backup efficiency, however, on the basis of large data volume, the backup speed sometimes cannot meet the actual requirement.
Disclosure of Invention
The application aims to provide an Internet data backup method, an Internet data backup device and a server.
Other features and advantages of the present application will be apparent from the following detailed description, or may be learned in part by the practice of the application.
According to an aspect of the embodiments of the present application, there is provided an internet data backup method, applied to a data backup server, the method including: acquiring internet data to be backed up, and mining a priority description carrier of the internet data to be backed up to obtain a data priority description carrier; the data priority description carrier is subjected to father data carrier mining to obtain father data carriers, and father priority analysis is carried out through the father data carriers to obtain data priority support coefficients corresponding to each piece of father priority information; carrying out sub-level data carrier mining on the data priority description carrier to obtain a sub-level data carrier, carrying out carrier integration processing on the parent data carrier and the sub-level data carrier to obtain a sub-level integrated carrier, and calling the sub-level integrated carrier to carry out sub-level priority analysis to obtain a data priority support coefficient corresponding to each piece of sub-level priority information, wherein the priority refinement index corresponding to the sub-level priority information is larger than the priority refinement index corresponding to the parent level priority information; obtaining data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father priority information and the data priority support coefficient corresponding to each son priority information, wherein the data priority comprises target father priority information and target son priority information corresponding to the target father priority information; and backing up the internet data to be backed up based on the data priority corresponding to the internet data to be backed up.
As an embodiment, the carrier integration processing is performed by a parent data carrier and a child data carrier to obtain a child integrated carrier, which includes: the parent data carrier and the child data carrier are subjected to tail-end operation to obtain a tail-end data carrier; performing gradient optimization carrier mining through the match tail data carrier and the sub-level data carrier to obtain a gradient optimization carrier, and performing normalization treatment on the gradient optimization carrier to obtain a sub-level integration carrier; wherein, carry on the optimization carrier of gradient through the data carrier of the armature and sub-level data carrier and excavate, get the optimization carrier of gradient, include: and carrying out embedded coding on the armature data carrier according to a preset gradient optimization variable to obtain an embedded coding carrier, and obtaining a carrier summation result of the embedded coding carrier and the sub-level data carrier to obtain the gradient optimization carrier.
As an implementation manner, obtaining the data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each parent priority information and the data priority support coefficient corresponding to each child priority information includes: acquiring data priority level information; integrating the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level through the priority level information of the data to obtain the combined support coefficients of each level; determining target level combination support coefficients in the level combination support coefficients, and determining data priority corresponding to the internet data to be backed up through the target level combination support coefficients; the method for integrating the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level through the data priority level information to obtain each level combination support coefficient comprises the following steps: screening out the data priority support coefficient corresponding to the current parent priority information from the data priority support coefficient corresponding to each parent priority information; acquiring current sub-level priority information corresponding to the current parent level priority information through the data priority level information, and determining a data priority support coefficient corresponding to the current sub-level priority information from the data priority support coefficients corresponding to the sub-level priority information; obtaining a multiplication result of a data priority support coefficient corresponding to the current parent priority information and a data priority support coefficient corresponding to the current child priority information to obtain a current level combination support coefficient; traversing the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level to obtain each level combination support coefficient.
As an embodiment, the method further comprises: loading the internet data to be backed up into a first target internet data backup network; the method comprises the steps of mining a priority description carrier of internet data to be backed up through a first target internet data backup network to obtain a data priority description carrier; the data priority description carrier is subjected to father data carrier mining through a first target Internet data backup network to obtain father data carriers, father priority analysis is carried out through the father data carriers, and data priority support coefficients corresponding to the father priority information are obtained; carrying out sub-level data carrier mining on the data priority description carrier through a first target Internet data backup network to obtain a sub-level data carrier, carrying out carrier integration processing on the parent data carrier and the sub-level data carrier to obtain a sub-level integrated carrier, and calling the sub-level integrated carrier to carry out sub-level priority analysis to obtain data priority support coefficients corresponding to each piece of sub-level priority information; the first target Internet data backup network comprises a priority description carrier mining module, a father priority analysis module and a son priority analysis module; loading the internet data to be backed up into a first target internet data backup network, comprising: loading the internet data to be backed up into a priority description carrier mining module to mine the priority description carrier, so as to obtain a data priority description carrier; loading the data priority description carrier into a father priority analysis module, mining the data priority description carrier through the father priority analysis module to obtain a father data carrier, and obtaining the data priority support coefficients corresponding to the father priority information through the father data carrier; loading the data priority description carrier and the father data carrier into a child priority analysis module, carrying out child data carrier mining on the data priority description carrier through the child priority analysis module to obtain a child data carrier, carrying out carrier integration processing on the father data carrier and the child data carrier to obtain a child integration carrier, and calling the child integration carrier to obtain a child priority information support coefficient to obtain a data priority support coefficient corresponding to each piece of child priority information; the sub-level priority analyzing module comprises a sub-level data carrier mining sub-module, a carrier integration processing sub-module and a sub-level priority analyzing sub-module; loading the data priority description carrier and the parent data carrier into a child priority resolution module, comprising: loading the data priority description carrier into a sub-data carrier mining sub-module to mine the sub-data carrier to obtain a sub-data carrier; loading the parent data carrier and the child data carrier into a carrier integration processing sub-module for carrier integration processing to obtain a child integration carrier; and loading the sub-level integrating carrier into a sub-level priority analysis sub-module to acquire sub-level priority information support coefficients, so as to acquire data priority support coefficients corresponding to each piece of sub-level priority information.
As an embodiment, the method further comprises: mining the data priority description carrier into a transition data carrier to obtain the transition data carrier; carrying out carrier integration treatment through a parent data carrier and a transition data carrier to obtain a transition integration carrier, and carrying out transition priority analysis through the transition integration carrier to obtain data priority support coefficients corresponding to each transition priority information; carrying out carrier integration treatment on the father data carrier, the transition integration carrier and the son data carrier to obtain a target son integration carrier, and calling the target son integration carrier to carry out son priority analysis to obtain a target data priority support coefficient corresponding to each piece of son priority information, wherein the priority refinement index corresponding to the son priority information is larger than the priority refinement index corresponding to the transition priority information, and the priority refinement index corresponding to the transition priority information is larger than the priority refinement index corresponding to the father priority information; obtaining target data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father-level priority information, the data priority support coefficient corresponding to each transition-level priority information and the target data priority support coefficient corresponding to each son-level priority information, wherein the target data priority comprises target father-level priority information, target transition-level priority information corresponding to the target father-level priority information and target son-level priority information corresponding to the target transition-level priority information; wherein, carry on the carrier integration treatment through father stage data carrier and transition stage data carrier, get the transition and integrate the carrier, include: the parent data carrier and the transition data carrier are subjected to tail linking to obtain a transition tail linking data carrier; performing gradient optimization carrier mining through the transition armature data carrier and the sub-level data carrier to obtain a transition gradient optimization carrier; normalizing the transition gradient optimization vector to obtain a transition integration vector; carrying out carrier integration treatment on the father data carrier, the transition integration carrier and the son data carrier to obtain a target son integration carrier, wherein the carrier comprises the following components: performing tail linking on the father data carrier, the transition integration carrier and the son data carrier to obtain a target tail linking data carrier; performing gradient optimization carrier mining through the target armature data carrier and the sub-level data carrier to obtain a target gradient optimization carrier; and (3) carrying out normalization treatment on the target gradient optimization vector to obtain the target sub-level integration vector.
As an embodiment, further comprising: loading the internet data to be backed up into a second target internet data backup network; the method comprises the steps that priority description carrier mining is conducted on internet data to be backed up through a second target internet data backup network, and a data priority description carrier is obtained; the data priority description carrier is subjected to father data carrier mining through a second target Internet data backup network to obtain father data carriers, father priority analysis is carried out through the father data carriers, and data priority support coefficients corresponding to the father priority information are obtained; the data priority description carrier is subjected to transition level data carrier mining through a second target Internet data backup network to obtain a transition level data carrier, carrier integration processing is carried out through a father level data carrier and the transition level data carrier to obtain a transition integration carrier, and transition level priority analysis is carried out through the transition integration carrier to obtain data priority support coefficients corresponding to each transition level priority information; carrying out carrier integration processing on a father data carrier, a transition integration carrier and a son data carrier through a second target internet data backup network to obtain a target son integration carrier, and calling the target son integration carrier to carry out son priority analysis to obtain target data priority support coefficients corresponding to each piece of son priority information; and obtaining target data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father-level priority information, the data priority support coefficient corresponding to each transition-level priority information and the target data priority support coefficient corresponding to each son-level priority information, wherein the target data priority comprises target father-level priority information, target transition-level priority information corresponding to the target father-level priority information and target son-level priority information corresponding to the target transition-level priority information.
As an embodiment, the method further comprises a training process of the internet data backup network, including the following steps: acquiring an Internet data sample and a corresponding Internet data priority indication; loading the internet data sample into a first basic internet data backup network, and mining a priority description carrier of the internet data sample through the first basic internet data backup network to obtain a sample data priority description carrier; the method comprises the steps of mining a sample data priority description carrier through a first basic internet data backup network to obtain a sample parent data carrier, and carrying out parent priority analysis through the sample parent data carrier to obtain sample data priority support coefficients corresponding to each piece of parent priority information; carrying out sub-level data carrier mining on the sample data priority description carrier through a first basic internet data backup network to obtain a sample sub-level data carrier, carrying out carrier integration processing on the sample parent data carrier and the sample sub-level data carrier to obtain a sample sub-level integration carrier, and calling the sample sub-level integration carrier to carry out sub-level priority analysis to obtain a sample data priority support coefficient corresponding to each piece of sub-level priority information, wherein the priority refinement index corresponding to the sub-level priority information is larger than the priority refinement index corresponding to the parent priority information; obtaining errors between sample data priority support coefficients corresponding to the priority information of each father level and father level priority information indications in the internet data priority indications to obtain father level priority information error information, and obtaining errors between sample data priority support coefficients corresponding to the priority information of each son level and the son level priority information indications in the internet data priority indications to obtain son level priority information error information; optimizing the first basic Internet data backup network through the parent priority information error information and the child priority information error information, and simultaneously returning to the step of obtaining the Internet data sample and the corresponding Internet data priority indication to repeat until the preset convergence requirement is met, so as to obtain the first target Internet data backup network.
As an embodiment, the method further comprises: loading the internet data sample into a second basic internet data backup network, and mining a priority description carrier of the internet data sample through the second basic internet data backup network to obtain a sample data priority description carrier; the sample data priority description carrier is subjected to father data carrier mining through a second basic Internet data backup network to obtain a sample father data carrier, and father priority analysis is performed through the sample father data carrier to obtain sample data priority support coefficients corresponding to each piece of father priority information; the sample data priority description carrier is subjected to transition level data carrier mining through a second basic internet data backup network to obtain a sample transition level data carrier, carrier integration processing is carried out through a sample father level data carrier and the sample transition level data carrier to obtain a sample transition integration carrier, transition level priority analysis is carried out through the sample transition integration carrier, and sample data priority support coefficients corresponding to each transition level priority information are obtained; carrying out carrier integration treatment on a sample father data carrier, a sample transition integration carrier and a sample son data carrier through a second basic internet data backup network to obtain a sample target son integration carrier, and calling the sample target son integration carrier to carry out son priority analysis to obtain a sample target data priority support coefficient corresponding to each son priority information, wherein the priority refinement index corresponding to the son priority information is larger than the priority refinement index corresponding to the transition priority information, and the priority refinement index corresponding to the transition priority information is larger than the priority refinement index corresponding to the father priority information; obtaining errors between sample data priority support coefficients corresponding to each piece of father-level priority information and father-level priority information indications in internet data priority indications, obtaining father-level priority information error information, obtaining errors between sample data priority support coefficients corresponding to each piece of transition priority information and transition priority information indications in the internet data priority indications, obtaining transition priority information error information, and obtaining errors between sample target data priority support coefficients corresponding to each piece of child priority information and the internet data priority indications, and obtaining child priority information target error information; optimizing the second basic internet data backup network through the parent priority information error information, the transition priority information error information and the child priority information target error information, and simultaneously returning to the step of obtaining the internet data sample and the corresponding internet data priority indication for repeating until the preset convergence requirement is met, so as to obtain the second target internet data backup network.
According to an aspect of the embodiments of the present application, there is provided a data backup apparatus, including: the data acquisition module is used for acquiring the internet data to be backed up, and carrying out priority description carrier mining on the internet data to be backed up to obtain a data priority description carrier; the carrier mining module is used for mining the data priority description carrier into a father data carrier to obtain the father data carrier, and performing father priority analysis through the father data carrier to obtain the data priority support coefficient corresponding to each father priority information; the carrier integration module is used for carrying out carrier mining on the data priority description carrier to obtain a sub-level data carrier, carrying out carrier integration processing on the parent data carrier and the sub-level data carrier to obtain a sub-level integration carrier, and calling the sub-level integration carrier to carry out sub-level priority analysis to obtain a data priority support coefficient corresponding to each piece of sub-level priority information, wherein the priority refinement index corresponding to the sub-level priority information is larger than the priority refinement index corresponding to the parent priority information; the priority determining module is used for obtaining the data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father priority information and the data priority support coefficient corresponding to each son priority information, wherein the data priority comprises target father priority information and target son priority information corresponding to the target father priority information; and the data backup module is used for backing up the internet data to be backed up based on the data priority corresponding to the internet data to be backed up.
According to an aspect of the embodiments of the present application, there is provided a data backup server, including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the above method via execution of the executable instructions.
The embodiment of the application at least comprises the following beneficial effects:
according to the internet data backup method, the server and the medium, the priority of the father level and the son level of the internet data to be backed up is divided, and the data with high priority can be backed up preferentially by carrying out data backup according to the priority, so that loss caused by emergency is reduced. In addition, in the process of dividing the data priority, the internet data backup network is adopted to conduct data priority identification and determination, so that the accuracy and the efficiency of data priority determination are improved, and the rationality and the efficiency of data backup are improved. In the debugging process of the internet data backup network, internet data samples are loaded into a first basic internet data backup network to carry out priority description carrier mining on the internet data samples to obtain sample data priority description carriers, father-level data carrier mining is carried out on the sample data priority description carriers to obtain sample father-level data carriers, father-level priority analysis is carried out on the sample father-level data carriers to obtain sample data priority support coefficients corresponding to each father-level priority information, son-level data carrier mining is carried out on the sample data priority description carriers to obtain sample son-level data carriers, carrier integration processing is carried out on the sample father-level data carriers and the sample son-level data carriers to obtain sample son-level integration carriers, and son-level priority analysis is carried out on the sample data priority integration carriers to obtain the sample data priority support coefficients corresponding to each son-level priority information. In the network debugging process, carrier integration processing is carried out based on a sample father-stage data carrier and a sample son-stage data carrier to obtain a sample son-stage integration carrier, then son-stage priority analysis is carried out, wherein a priority refinement index corresponding to son-stage priority information is larger than a priority refinement index corresponding to father-stage priority information, based on the priority refinement index, the level composition of priority can be learned in the network debugging process, father-stage priority information error information and son-stage priority information error information can be obtained, optimized debugging is carried out based on the father-stage priority information error information and the son-stage priority information error information until a preset convergence requirement is met, and a first target Internet data backup network is obtained, so that the precision of the first target Internet data backup network is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application. It is apparent that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
Fig. 1 is a flowchart of an internet data backup method provided in an embodiment of the present application.
Fig. 2 is a schematic functional module architecture of a data backup device according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a data backup server according to an embodiment of the present application.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments may be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art.
Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the present application. One skilled in the relevant art will recognize, however, that the aspects of the application can be practiced without one or more of the specific details, or with other methods, components, devices, steps, etc. In other instances, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the application.
The block diagrams depicted in the figures are merely functional entities and do not necessarily correspond to physically separate entities. That is, the functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
The flow diagrams depicted in the figures are exemplary only, and do not necessarily include all of the elements and operations/steps, nor must they be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the order of actual execution may be changed according to actual situations.
According to the internet data backup method, when the internet data is backed up, the internet data is firstly subjected to priority classification, then is backed up according to the priority, for example, is sequentially backed up according to the priority, wherein when the priority classification is carried out, one upper level data possibly comprises a plurality of lower level data in consideration of the multi-level characteristic of the data, so that the priority of the internet data is continuously mined and classified into multiple priority levels, and the reasonable backup of the data is conveniently carried out according to the fine-granularity priority levels. Specifically, please refer to fig. 1, which is a flowchart of an internet data backup method provided in an embodiment of the present application, the internet data backup method provided in the embodiment of the present application includes the following steps:
s100: and acquiring the internet data to be backed up, and mining the priority description carrier of the internet data to be backed up to obtain the data priority description carrier.
In this embodiment of the present application, the internet data to be backed up may be internet data such as internet office information, internet business information (e.g. consumer data in e-commerce service, customer relationship data, etc.), internet financial private information, etc. Internet data may be received based on multiple databases or storage systems, for example, transaction data may be stored using the general relational databases MySQL and Oracle. In MySQL, the internet data to be backed up includes a transaction log and metadata, and history modification information related to the execution process of the transaction of the internet data is recorded in the transaction log, specifically, a log file includes a plurality of transaction entries, each transaction entry is a change carrier (change vector), and represents change information of a data block in a database. Metadata is data describing data, such as attribute information (storage location, history, record, etc.) of the data, and data priority description bearers characterize priority information of the internet data to be backed up.
S200: and carrying out father data carrier mining on the data priority description carrier to obtain a father data carrier, and carrying out father priority analysis on the data priority description carrier through the father data carrier to obtain data priority support coefficients corresponding to each piece of father priority information.
In this embodiment, for different data, there may be different backup priorities in the backup process, targeted backup is performed according to the priority of the information (for example, according to the priority order, data with a high priority is backed up first, and then data with a low priority is backed up), the priority setting may be classified into different classifications, and the different classifications may be further classified into different priorities, for example, priority classifications a and B, and priority classification a is further classified into priorities A1, A2, and A3. In step S200, the parent level, i.e., the level corresponding to the lowest level of refinement index (classified level granularity) in the data priority, and the data interval corresponding to the lowest level of refinement index in the data priority is the largest, more levels may be derived.
In the embodiment of the application, the data priorities are set into different grades in advance, and the father grade is the grade of the lowest priority in the data priorities. In particular, the data priority refinement index of the data priority hierarchy is smaller for a hierarchy closer to the parent and larger for a hierarchy. Different data priority classifications are formed based on different parent priority information, and each parent priority information corresponds to a data priority classification. The parent data carrier is vector information mined based on a DNN (deep neural network) model corresponding to the parent level and is used for identifying the data priority corresponding to the parent level. The father-level priority information is the data priority corresponding to the father-level in the data priority grading, the data priority support coefficient represents the certainty that the corresponding priority is the corresponding priority of the internet data to be backed up, and the larger the data priority support coefficient is, the larger the probability that the corresponding priority is the corresponding priority of the internet data to be backed up is. The data priority support coefficient corresponding to the parent priority information represents the certainty that the corresponding parent priority information is the corresponding priority of the internet data to be backed up, and the larger the data priority support coefficient is, the larger the probability that the corresponding priority of the internet data to be backed up is the parent priority information is. For example, the depth features of the data priority description carrier are mined based on DNNs corresponding to the father level to obtain the father level data carrier, and then the father level priority analysis is performed on the father level data carrier based on the father level priority analysis variables to obtain the data priority support coefficients corresponding to the priority information of each father level, wherein the father level priority analysis is, for example, a two-class, three-class or multi-class task, and the priority information of the father level is multiple.
S300: the data priority description carrier is subjected to sub-level data carrier mining to obtain a sub-level data carrier, the sub-level data carrier is subjected to carrier integration processing through the father-level data carrier and the sub-level data carrier to obtain a sub-level integrated carrier, the sub-level integrated carrier is called to carry out sub-level priority analysis to obtain data priority support coefficients corresponding to all sub-level priority information, and the priority refinement index corresponding to the sub-level priority information is larger than the priority refinement index corresponding to the father-level priority information.
In this embodiment of the present application, the sub-level is a level corresponding to a priority with a maximum refinement index among the data priorities, and a data interval corresponding to a priority with a maximum refinement index among the data priorities is the smallest. The sub-level data carrier is vector information mined based on DNNs corresponding to sub-levels, and the DNNs corresponding to different hierarchical configurations of the data priority levels are used for completing mining of depth features, and is used for identifying the data priority levels corresponding to the sub-levels. The child level integration carrier is obtained by fusing the parent level data carrier and the child level data carrier, and parent level information can be considered when the priority of the child level is determined, so that the definition of the priority of the child level is more accurate. The sub-level priority information is the data priority corresponding to the data priority grading sub-level, the priority refinement index represents the proximity of the data priority to the information, the larger the priority refinement index is, the finer the data priority granularity is, the higher the proximity of the information is, and in the data priority grading, the priority refinement index of the same grading data priority is the same. The priority refinement index corresponding to the child priority information is larger than the priority refinement index corresponding to the parent priority information in the parent level, and the data priority support coefficient corresponding to the child priority information represents the certainty that the corresponding child priority information is the corresponding priority of the internet data to be backed up, and the larger the value is, the larger the probability that the corresponding priority of the internet data to be backed up is the child priority information is. For example, the data priority description carrier is mined based on the DNN of the child level to obtain the child level data carrier, then the parent level data carrier and the child level data carrier are subjected to end-to-end (end-to-end splicing) to obtain the end-to-end data carrier, then the end-to-end data carrier is subjected to carrier mining (feature mining) to obtain the child level integration carrier, and then the child level integration carrier is subjected to child level priority analysis based on the child level priority analysis variable to obtain the data priority support coefficient corresponding to each piece of child level priority information.
S400: and obtaining the data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father priority information and the data priority support coefficient corresponding to each son priority information, wherein the data priority comprises target father priority information and target son priority information corresponding to the target father priority information.
In this embodiment of the present application, the target parent priority information is a priority of a parent in a data priority corresponding to the internet data to be backed up, the target child priority information is a priority of a child in the data priority corresponding to the internet data to be backed up, and the target parent priority information and the target child priority information are priorities in the data priority of the same composition architecture, in other words, a priority of a bottom layer position corresponding to a branching position of the target child priority information is the target parent priority information. In practical application, based on the parent priority information which is classified at the bottom position and corresponds to the branch position in advance, the branch of the child priority information is obtained, the multiplication result of the data priority support coefficient corresponding to the parent priority information and the data priority support coefficient corresponding to the child priority information on each branch is obtained, the data priority support coefficient of each branch is obtained, the branch of the maximum data priority support coefficient is screened, the parent priority information and the child priority information which are included in the branch of the maximum data priority support coefficient are determined as the target parent priority information and the target child priority information corresponding to the target parent priority information, and thus the data priority corresponding to the internet data to be backed up is obtained. As an implementation mode, parent priority information corresponding to the maximum data priority support coefficient can be directly screened out from the data priority support coefficients corresponding to each parent priority information to obtain target parent priority information, child priority information corresponding to the target parent priority information is obtained based on the data priority classification determined in advance, and child priority information corresponding to the maximum data priority support coefficient is screened out from the data priority support coefficients of all the child priority information corresponding to the target parent priority information to obtain target child priority information.
S500: and backing up the internet data to be backed up based on the data priority corresponding to the internet data to be backed up.
For example, the internet data to be backed up is backed up according to the sequence of the data priority, for example, the internet data with high data priority is backed up preferentially.
According to the embodiment of the application, the data priority description carrier is subjected to father data carrier mining to obtain the father data carrier, and then the father priority analysis is performed to obtain the data priority support coefficient corresponding to each piece of father priority information. The data priority description carrier is subjected to sub-level data carrier mining to obtain a sub-level data carrier, the parent data carrier and the sub-level data carrier are subjected to carrier integration processing to obtain a sub-level integration carrier, then sub-level priority analysis is carried out based on the sub-level integration carrier to obtain data priority support coefficients corresponding to all sub-level priority information, priority determination can be carried out on the sub-level based on depth information of the parent level, because priority refinement indexes corresponding to the sub-level priority information are larger than priority refinement indexes corresponding to the parent level priority information, the hierarchical architecture of the priority is comprehensively measured, the data priority support coefficients corresponding to all the obtained sub-level priority information are more accurate, then the data priority corresponding to the internet data to be backed up is obtained based on the data priority support coefficients corresponding to all the parent level priority information and the data priority support coefficients corresponding to all the sub-level priority information, and the data priority comprises target parent level priority information and target sub-level priority information corresponding to the target parent level priority information, so that the accuracy of the data priority is higher, and the accuracy of priority determination is further improved.
In one embodiment, in S300, the carrier integration process is performed by using the parent data carrier and the child data carrier to obtain a child level integrated carrier, which may specifically include: the parent data carrier and the child data carrier are subjected to tail-end operation to obtain a tail-end data carrier; and carrying out gradient optimization carrier mining through the armature data carrier and the sub-level data carrier to obtain a gradient optimization carrier, and carrying out normalization treatment on the gradient optimization carrier to obtain the sub-level integration carrier.
The data carrier of the armature is obtained by taking the vector of the parent data carrier and the vector of the child data carrier as the head and the vector of the child data carrier as the tail, for example, the vector of the parent data carrier is taken as the tail to finish the armature, so as to obtain the data carrier of the armature, or the vector of the parent data carrier is taken as the tail, and the vector of the child data carrier is taken as the head to carry out armature, so as to obtain the data carrier of the armature. Gradient optimization vector is vector information mined by using ResNet, and ResNet can reduce gradient disappearance caused by depth increment in DNN. For example, the parent data carrier and the child data carrier are subjected to tail-joining to obtain a tail-joining data carrier, gradient optimization carrier mining is carried out on the tail-joining data carrier and the child data carrier based on ResNet to obtain a gradient optimization carrier, and a normalization processing operator is called to normalize the gradient optimization carrier to obtain the child integration carrier. The normalization processing operator limits the gradient optimization carrier to a reasonable data interval to complete data standardization so as to prevent gradient problems.
As an embodiment, gradient optimization carrier mining is performed by the armature data carrier and the sub-level data carrier to obtain a gradient optimization carrier, which may specifically include: and carrying out embedded coding on the armature data carrier according to a preset gradient optimization variable (residual coefficient) to obtain an embedded coding carrier, and obtaining a carrier summation result of the embedded coding carrier and the sub-level data carrier to obtain the gradient optimization carrier. The preset gradient optimization variables are variables which are debugged in advance and used for carrying out residual operation, and specifically comprise eccentric variables (variables representing weighted indexes) and adjusting variables (variables representing offset indexes). The embedded encoding carrier is a characterization vector obtained by embedded encoding the armature data carrier, and can be realized based on linear projection. For example, weighting the armature data carrier according to an eccentric variable in a preset gradient optimization variable, obtaining a summation result of the weighted armature data carrier and an adjusting variable of the preset gradient optimization variable, obtaining an embedded encoding carrier, and finally summing the embedded encoding carrier and the sub-data carrier to obtain the gradient optimization carrier. The parent data carrier and the child data carrier are adopted to carry out tail connection to obtain a tail connection data carrier, gradient optimization carrier excavation is carried out through the tail connection data carrier and the child data carrier to obtain a gradient optimization carrier, and the gradient optimization carrier is subjected to normalization treatment to obtain a child integration carrier, so that the obtained child integration carrier is higher in precision, and meanwhile gradient disappearance is reduced.
As an implementation manner, in S400, obtaining the data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each parent priority information and the data priority support coefficient corresponding to each child priority information may specifically include:
s410: and acquiring data priority level information.
S420: and integrating the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level through the data priority level information to obtain each level combination support coefficient.
The data priority level information is predetermined and is used for describing the classification of each data priority, and different data priorities can correspond to the same data priority classification or different data priority classifications. For example, the data priority levels include A, B, C, the data priority levels of a include the data priority levels of B, and the data priority levels of B include the data priority levels of C. The data priority level information is obtained based on, for example, a data priority hierarchy of a hierarchy. The level combination support coefficient represents the probability that the data priority corresponding to the internet data to be backed up is the data priority included in the hierarchical branch. The branch line may be a line from the parent priority information to the child priority information in the data priority hierarchy, for example, a line from the data priority of the underlying location to the data priority of the branch location is a branch line. For example, data priority level information is obtained, and data priority support coefficients corresponding to the same hierarchical branch line in the data priority support coefficients corresponding to the parent priority information and the child priority information are integrated based on the data priority level information to obtain each level combination support coefficient.
S430: and determining a target level combination support coefficient from the level combination support coefficients, and determining the data priority corresponding to the internet data to be backed up through the target level combination support coefficient.
For example, by comparing the values of the respective level combination support coefficients, the maximum level combination support coefficient is determined as the target level combination support coefficient, and the parent level priority information and the child level priority information in the hierarchical branch line corresponding to the target level combination support coefficient are determined as the data priority corresponding to the internet data to be backed up. Based on the data priority level information, integrating the data priority support coefficient corresponding to the priority information of each father level and the data priority support coefficient corresponding to the priority information of each son level to obtain the combined support coefficient of each level. And determining a target level combination support coefficient from the level combination support coefficients, and determining the data priority corresponding to the internet data to be backed up through the target level combination support coefficient, so that the acquired data priority corresponding to the internet data to be backed up can contain an accurate hierarchical structure, and the situation that father-level priority information and son-level priority information in the data priority do not correspond to the same hierarchical branch line is prevented.
As an implementation manner, in S420, integrating, by the data priority level information, the data priority support coefficient corresponding to each parent level priority information and the data priority support coefficient corresponding to each child level priority information to obtain each level combination support coefficient, which may specifically include:
s421: and screening the data priority support coefficient corresponding to the current parent priority information from the data priority support coefficient corresponding to each parent priority information.
S422: acquiring current sub-level priority information corresponding to the current parent level priority information through the data priority level information, and determining a data priority support coefficient corresponding to the current sub-level priority information from the data priority support coefficients corresponding to the sub-level priority information.
In the step, the current parent priority information is the parent priority information which is required to be integrated and acquired currently, and the current child priority information is the child priority information which is required to be integrated and acquired with the parent priority information in all the child priority information corresponding to the current parent priority information. For example, the current parent priority information may be screened out from the parent priority information one by one, and the data priority support coefficient corresponding to the current parent priority information may be obtained from the data priority support coefficients corresponding to the parent priority information. And then, acquiring all the sub-level priority information contained in the current parent level priority information based on the data priority level information, determining the sub-level priority information as the current sub-level priority information one by one, and acquiring the data priority support coefficient corresponding to the current sub-level priority information from the data priority support coefficients corresponding to the sub-level priority information.
S423: and obtaining a multiplication result of the data priority support coefficient corresponding to the current parent priority information and the data priority support coefficient corresponding to the current child priority information to obtain the current level combination support coefficient.
S424: traversing the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level to obtain each level combination support coefficient.
The current level combination support coefficient is the probability that the current parent level priority information and the current child level priority information are the data priority corresponding to the internet data to be backed up. For example, the data priority support coefficient corresponding to the current parent priority information and the data priority support coefficient corresponding to the current child priority information are integrated to obtain a current level combination support coefficient, and the data priority support coefficient corresponding to the parent priority information and the data priority support coefficient corresponding to the child priority information on each hierarchical branch line are integrated to obtain each level combination support coefficient.
For example, the data priority level information includes parent level priority information X and parent level priority information Y, each of which includes three child level priority information, respectively child level priority information X a Sub-level priority information X b Sub-level priority information X c Sub-level priority information Y a Sub-level priority information Y b Sub-level priority information Y c . Then, the classification branch line comprises six lines, X and X can be combined a Determining to acquire the current parent priority information and the current child priority information, and traversing all the parent priority information and the child priority information until the acquisition is completed c When the corresponding level combination support coefficient is obtained, six level combination support coefficients are obtained, the largest level combination support coefficient is screened out from the six level combination support coefficients, and the data priority in the hierarchical branch line corresponding to the largest level combination support coefficient is determined as the data priority corresponding to the internet data to be backed up. Based on the above process, acquiring current sub-level priority information corresponding to the current parent level priority information through the data priority level information, and acquiring a multiplication result of the data priority support coefficient corresponding to the current parent level priority information and the data priority support coefficient corresponding to the current sub-level priority information to obtain a current level combination support coefficient; and traversing the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level to obtain each level combination support coefficient, so that the accuracy of each level combination support coefficient is improved.
As an implementation manner, the internet data backup method provided by the application further includes the following steps:
s210: and loading the internet data to be backed up into the first target internet data backup network.
S220: and mining the priority description carrier of the internet data to be backed up through the first target internet data backup network to obtain the data priority description carrier.
The target internet data backup network is a network which is debugged in advance and is used for dividing information into multiple levels of priority, and the first target internet data backup network is a target internet data backup network comprising father level priority analysis and son level priority analysis. For example, an internet debug sample is obtained, a first target internet data backup network is obtained by using the internet debug sample, and then the first target internet data backup network is configured. In the backup task, the acquired internet data to be backed up is determined as the loading data of the first target internet data backup network, the first target internet data backup network performs priority description carrier mining on the internet data to be backed up to obtain a data priority description carrier, specifically, mining can be performed on the basis of a neural network of the data priority description carrier, for example, data disassembly is performed on the internet data to be backed up to obtain a plurality of data blocks, quantization results of the data blocks are acquired, and the data priority description carrier is obtained on the basis of the quantization results.
S230: the data priority description carrier is subjected to father data carrier mining through a first target Internet data backup network to obtain father data carriers, father priority analysis is carried out through the father data carriers, and data priority support coefficients corresponding to the father priority information are obtained;
s240: the data priority description carrier is subjected to sub-level data carrier mining through the first target Internet data backup network to obtain a sub-level data carrier, carrier integration processing is carried out through the father data carrier and the sub-level data carrier to obtain a sub-level integration carrier, and the sub-level integration carrier is called to carry out sub-level priority analysis to obtain data priority support coefficients corresponding to each piece of sub-level priority information.
For example, after the data priority description carrier is obtained based on the mining of the first target internet data backup network, the parent priority analysis and the child priority analysis are performed simultaneously. The data priority description carrier can be subjected to father-stage data carrier mining to obtain father-stage data carriers, father-stage data carrier analysis is performed through the father-stage data carriers to obtain data priority support coefficients corresponding to the information of each father-stage data, then the data priority description carrier is subjected to son-stage data carrier mining to obtain son-stage data carriers, carrier integration processing is performed through the father-stage data carriers and the son-stage data carriers to obtain son-stage integrated carriers, and son-stage priority analysis is performed through the calling son-stage integrated carriers to obtain the data priority support coefficients corresponding to the information of each son-stage data priority. Based on the above process, the data priority support coefficient corresponding to each parent priority information and the data priority support coefficient corresponding to each child priority information are obtained, and the speed of priority determination is improved.
As one embodiment, the first target internet data backup network includes a priority description carrier mining module, a parent priority resolution module, and a child priority resolution module. In S210, loading the internet data to be backed up into the first target internet data backup network may specifically include:
s211: and loading the internet data to be backed up into a priority description carrier mining module to mine the priority description carrier, so as to obtain the data priority description carrier.
S212: the data priority description carrier is loaded into a father priority analysis module, the father data carrier is mined by the father priority analysis module, the father data carrier is obtained, and the father data carrier is used for obtaining the father priority information support coefficient, so that the data priority support coefficient corresponding to each father priority information is obtained.
S213: the data priority description carrier and the father data carrier are loaded into a child priority analysis module, child data carrier mining is carried out on the data priority description carrier through the child priority analysis module, the child data carrier is obtained, carrier integration processing is carried out on the father data carrier and the child data carrier, the child level integration carrier is obtained, and the child level integration carrier is called to obtain child level priority information support coefficients, so that the data priority support coefficients corresponding to all the child level priority information are obtained.
The priority description carrier mining module may be DNN for performing priority description carrier mining, and the priority description carrier mining module is a backbone network for performing priority identification, for example, a network such as CNN, BERT, LSTM, transformer. The parent priority analyzing module is used for carrying out a sub-network of priority determination on each piece of pre-determined parent priority information, the sub-priority analyzing module is used for carrying out a sub-network of priority determination on each piece of pre-determined sub-priority information, and the parent priority analyzing module and the sub-priority analyzing module are constructed based on DNN, for example. For example, the priority description carrier mining module is used for mining the priority description carrier, and then the data priority description carrier is loaded into two sub-networks, namely the parent priority analysis module and the child priority analysis module, and the child priority analysis module also obtains the parent data carrier extracted by the parent priority analysis module when determining the priority. And carrying out carrier integration treatment on the parent data carrier and the child data carrier to obtain a child level integrated carrier, and carrying out priority determination based on the child level integrated carrier. Based on the above process, after the priority information is extracted by the priority description carrier mining module, priority determination is performed based on the parent priority analysis module and the child priority analysis module, and the speed and the accuracy of the priority determination are increased.
As an embodiment, the sub-level priority parsing module includes a sub-level data carrier mining sub-module, a carrier integration processing sub-module, and a sub-level priority parsing sub-module. In S213, loading the data priority description carrier and the parent data carrier into the child priority parsing module may specifically include: the data priority description carrier is loaded into a sub-level data carrier mining sub-module to conduct sub-level data carrier mining to obtain a sub-level data carrier, the father-level data carrier and the sub-level data carrier are loaded into a carrier integration processing sub-module to conduct carrier integration processing to obtain a sub-level integration carrier, and the sub-level integration carrier is loaded into a sub-level priority analysis sub-module to conduct sub-level priority information support coefficient acquisition to obtain data priority support coefficients corresponding to all sub-level priority information. The sub-level data carrier mining submodule is used for mining sub-level data carriers, and the carrier integration processing submodule is used for integrating characteristic information of different grades. The sub-level priority analysis sub-module is a network for carrying out priority determination on each sub-level priority information. For example, the sub-level data carrier is mined based on the sub-level data carrier mining sub-module, the sub-level data carrier and the father level data carrier are integrated through the carrier integration processing sub-module, the sub-level integration carrier is obtained, the sub-level priority data priority is determined according to the sub-level priority analysis sub-module, and therefore accuracy of the data priority support coefficient corresponding to each piece of obtained sub-level priority information is improved.
Loading the data priority description carrier into the parent data carrier mining network Model1 to obtain the parent data carrier, for example, completing depth feature mining based on the following formula:
G 1 =A 1 ·G c +f 1
G 1 for parent data carrier G c The carrier is described for data priority. A is that 1 And f 1 Is the network parameter (eccentric weight, adjustment bias coefficient) of Model 1. And then, determining the priority based on the father-stage data carrier to obtain the data priority support coefficient corresponding to each father-stage priority information. The priority determination is made, for example, based on the following formula.
G 2 =sigmoid（A 2 ·G c +f 2 ）·P 1
P 1 And loading the data priority description carrier into a sub-data carrier mining network Model2 to obtain an output sub-data carrier, for example, performing depth feature mining based on the following formula to obtain the sub-data carrier:
G 2 =sigmoid（A 2 ·G c +f 2 ）
G 2 for sub-level data carrier G c The carrier is described for data priority. A is that 2 And f 2 And (3) carrying out carrier integration treatment on the parent data carrier and the child data carrier to obtain the child level integrated carrier as the network parameter of the Model 2. The carrier integration process is performed based on the following formula,to obtain the sub-level integration vector.
G 3 =concat(G 1 ，G 2 ）
G m =BN（G 2 +ReLu（Am·G 3 +f m ））
G 3 For the armature vector, G m For subset integration vectors, A m Is the eccentric variable during splicing, f m And BN is normalized and is an adjusting variable during splicing. And then, carrying out priority determination based on the sub-level integrating carrier to obtain the data priority support coefficient corresponding to each sub-level priority information. The priority determination is made, for example, based on the following formula:
P 2 =SM（A p2 ·Gm+f p2 ）
P 2 and supporting coefficients for the data priority corresponding to each piece of sub-level priority information. The SM determines the network for the priority of the sub-level, e.g. activates the network softmax. A is that p2 For eccentric variable in sub-level priority resolution, f p2 Is the adjustment variable in sub-level priority resolution. And finally, obtaining a data priority support coefficient corresponding to each piece of child priority information and a data priority support coefficient corresponding to each piece of father priority information, and determining the data priority corresponding to the internet data to be backed up, wherein the data priority comprises target father priority information and target child priority information corresponding to the target father priority information, so that the accuracy of the data priority is improved.
As an implementation manner, the internet data backup method provided by the embodiment of the application may further include:
s310: and excavating the transition data carrier by the data priority description carrier to obtain the transition data carrier.
S320: and carrying out carrier integration processing by the parent data carrier and the transition data carrier to obtain a transition integration carrier, and carrying out transition priority analysis by the transition integration carrier to obtain data priority support coefficients corresponding to each transition priority information.
The transition level is a hierarchy of data priorities determined in advance, and is located between a parent level and a child level, the transition level may include a plurality of transition level data carriers, the transition level data carriers are vector information mined based on DNNs corresponding to the transition level, the transition level data carriers are used for identifying the data priorities corresponding to the transition level, and the transition integration carrier is carrier information obtained by integrating the parent level data carrier and the transition level data carrier, so that when the priority of the transition level is determined, carrier information of a previous layer can be integrated. When the transition level is provided, the upper level is a father level, and when the transition level is provided for a plurality of times, the upper level is a father level or a previous transition level, and the transition level priority information is the data priority classified between the father level and the son level in the data priority classification. For example, the transition level data carrier is obtained by mining depth features of the data priority description carrier based on transition level depth feature mining variables. And then, the father-stage data carrier and the transition-stage data carrier are subjected to tail connection to obtain a tail connection carrier, and the carrier after tail connection is subjected to characteristic mining to obtain a transition integration carrier. And then, carrying out transition priority analysis through a transition integration carrier to obtain data priority support coefficients corresponding to the transition priority information.
S330: carrying out carrier integration processing on the father data carrier, the transition integration carrier and the son data carrier to obtain a target son integration carrier, and calling the target son integration carrier to carry out son priority analysis to obtain target data priority support coefficients corresponding to all son priority information, wherein the priority refinement index corresponding to the son priority information is larger than the priority refinement index corresponding to the transition priority information, and the priority refinement index corresponding to the transition priority information is larger than the priority refinement index corresponding to the father priority information.
When the target sub-level integrating carrier is a carrier obtained after integrating the father-level data carrier, the transition integrating carrier and the sub-level data carrier, the target sub-level integrating carrier comprises characteristic information for identifying the father-level, the transition and the sub-level, and the priority refinement index corresponding to the transition priority information is smaller than the priority refinement index corresponding to the sub-level priority information and is larger than the priority refinement index corresponding to the father-level priority information, in other words, the priority refinement index corresponding to the transition priority information is between the priority refinement index of the father-level priority information and the priority refinement index of the sub-level priority information.
For example, after determining the text priority of the transition level, determining the text priority of the sub-level, and performing carrier integration processing through the parent data carrier, the transition integration carrier and the sub-level data carrier to obtain a target sub-level integration carrier corresponding to the sub-level. The method comprises the steps of carrying out tail joining on a father stage data carrier, a transition integration carrier and a son stage data carrier, carrying out gradient optimization carrier mining on characteristics after tail joining, and carrying out normalization treatment on the gradient optimization carrier to obtain a target son stage integration carrier. And then carrying out priority determination of the child level based on the target child level integrating carrier, and carrying out priority determination of the child level based on the target child level integrating carrier, wherein the child level integrates the characteristics of the parent level and the transition level for carrying out priority when carrying out priority determination, so that the resolution and identification precision of the child level is improved.
S340: and obtaining target data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father-level priority information, the data priority support coefficient corresponding to each transition-level priority information and the target data priority support coefficient corresponding to each son-level priority information, wherein the target data priority comprises target father-level priority information, target transition-level priority information corresponding to the target father-level priority information and target son-level priority information corresponding to the target transition-level priority information.
The target transition level priority information is the priority of the transition level of the data priority class corresponding to the internet data to be backed up. For example, based on the data priority determined in advance, determining transition priority information from parent priority information corresponding to a bottom layer position to intermediate position, and then to branches of child priority information corresponding to a branch position, obtaining data priority support coefficients corresponding to the parent priority information and data priority support coefficients corresponding to the transition priority information on each branch, obtaining multiplication results of the data priority support coefficients corresponding to the child priority information and the data priority support coefficients corresponding to the child priority information, obtaining the data priority support coefficients of each branch, screening out the branch with the largest data priority support coefficient, and determining the parent priority information, the transition priority information and the child priority information included in the branch with the largest data priority support coefficient as target parent priority information, target transition priority information corresponding to the target parent priority information and target child priority information corresponding to the target transition priority information, thereby obtaining the target data priority corresponding to the internet data to be backed up.
As one implementation mode, the parent priority information corresponding to the maximum data priority support coefficient can be directly screened out from the data priority support coefficients corresponding to the parent priority information to obtain the target parent priority information, each piece of transition priority information corresponding to the target parent priority information is obtained based on the data priority classification determined in advance, and then the transition priority information corresponding to the maximum data priority support coefficient is screened out from the data priority support coefficients of each piece of transition priority information corresponding to the target parent priority information to obtain the target transition priority information. And then, each piece of sub-level priority information corresponding to the target transition level priority information is obtained based on the data priority level which is determined in advance, and the sub-level priority information corresponding to the maximum data priority support coefficient is screened out from the data priority support coefficients of each piece of sub-level priority information corresponding to the target father level priority information, so that the target sub-level priority information is obtained. Based on the above process, the priority of the transition level is determined by the data priority description carrier, so that the target data priority comprises target father level priority information, target transition level priority information corresponding to the target father level priority information and target son level priority information corresponding to the target transition level priority information, and the accuracy of the target data priority is improved.
As an embodiment, the carrier integration processing is performed on the parent data carrier and the transition data carrier in S320 to obtain a transition integration carrier, which may specifically include: the parent data carrier and the transition data carrier are subjected to tail linking to obtain a transition tail linking data carrier; and carrying out gradient optimization carrier mining through the transition armature data carrier and the sub-level data carrier to obtain a transition gradient optimization carrier, and carrying out normalization treatment on the transition gradient optimization carrier to obtain a transition integration carrier. The transition-end data carrier is obtained by end-connecting a parent data carrier and a transition data carrier, and is obtained by the transition stage during carrier integration processing. The transition gradient optimization carrier is a gradient optimization carrier excavated by the transition stage when the carrier integration treatment is carried out, and the transition integration carrier is an integration characteristic excavated by the transition stage when the carrier integration treatment is carried out. For example, the parent data carrier and the transition data carrier are subjected to head-tail connection to obtain a transition connection data carrier, then gradient optimization carrier mining is carried out on the transition connection data carrier and the transition data carrier based on ResNet to obtain a transition gradient optimization carrier, and a normalization processing operator is called to normalize the transition gradient optimization carrier to obtain a transition integration carrier, wherein the normalization processing operator is used for limiting the transition gradient optimization carrier to a reasonable interval so as to prevent gradient problems. And when the gradient optimization carrier is excavated, the transition armature data carrier is subjected to linear projection to obtain an embedded coding carrier of the transition stage, and then a carrier summation result of the embedded coding carrier of the transition stage and the transition stage data carrier is obtained to obtain the transition gradient optimization carrier. Based on the above procedure, the parent data carrier is integrated with the depth features of the transition level upon the priority determination of the transition level, and then the priority determination of the transition level is performed such that the priority of the transition level is more accurate.
As an embodiment, in S330, the parent data carrier, the transition integration carrier and the child data carrier are subjected to carrier integration treatment to obtain a target child level integration carrier, which may specifically include: performing tail linking on the father data carrier, the transition integration carrier and the son data carrier to obtain a target tail linking data carrier; and carrying out gradient optimization carrier mining through the target armature data carrier and the sub-level data carrier to obtain a target gradient optimization carrier, and carrying out normalization treatment on the target gradient optimization carrier to obtain a target sub-level integration carrier.
The target mouthpiece data carrier is a mouthpiece data carrier obtained by subjecting a parent data carrier, a transitional integration carrier, and a child data carrier to mouthpiece processing when the transitional stage is included, in other words, the target mouthpiece data carrier is a mouthpiece data carrier obtained by subjecting the child stage to carrier integration processing. The target gradient optimization carrier is a gradient optimization carrier with a transition level time sub-level excavated when the carrier integration treatment is carried out, and the target sub-level integration carrier is an integration characteristic with a transition level time sub-level excavated when the carrier integration treatment is carried out. For example, the parent data carrier, the transition level depth carrier summation result, and the child data carrier are subjected to tail-end, so as to obtain the target tail-end data carrier. And then carrying out gradient optimization carrier mining on the target armature data carrier and the sub-level data carrier according to the ResNet to obtain a target gradient optimization carrier, and carrying out normalization processing on the target gradient optimization carrier by a normalization processing operator to obtain a target sub-level integration carrier. Based on the above process, all grading depth features are integrated through sub-level priority analysis, so that a target sub-level integrated carrier is obtained, the accuracy of the obtained target sub-level integrated carrier is improved, and then the sub-level priority is determined according to the target sub-level integrated carrier, so that the sub-level priority identification accuracy is higher.
As an embodiment, the internet data backup method may further include:
s10: and loading the internet data to be backed up into a second target internet data backup network, and mining the priority description carrier of the internet data to be backed up through the second target internet data backup network to obtain a data priority description carrier.
S20: and excavating the data priority description carrier through the second target Internet data backup network to obtain a father data carrier, and analyzing the father priority through the father data carrier to obtain data priority support coefficients corresponding to the information of each father priority.
The second target internet data backup network is a target internet data backup network comprising a parent priority resolution, a transition priority resolution and a child priority resolution. As one embodiment, a parent priority resolution, a plurality of transition priority resolutions, and a child priority resolution may be included in the target internet data backup network. The identification discrimination parameters (classification parameters) for carrying out father priority resolution in the second target internet data backup network are consistent with the identification discrimination parameters for carrying out father priority resolution in the first target internet data backup network, when the data priority of the father level determined in advance is the same, the identification discrimination parameters for carrying out father priority resolution in the second target internet data backup network are consistent with the identification discrimination parameters for carrying out father priority resolution in the first target internet data backup network, and when the data priority of the father level determined in advance is different, the identification discrimination parameters for carrying out father priority resolution in the second target internet data backup network are inconsistent with the identification discrimination parameters for carrying out father priority resolution in the second target internet data backup network. For example, the priority of the father level is determined through the second target internet data backup network, in other words, the father level data carrier is mined based on the father level depth feature mining network, the father level priority analysis is performed according to the father level data carrier based on the father level classification and identification network, and the data priority support coefficient corresponding to each father level priority information is obtained.
S30: and mining the data priority description carrier through a second target Internet data backup network to obtain a transition data carrier, carrying out carrier integration processing through the father data carrier and the transition data carrier to obtain a transition integration carrier, and carrying out transition priority analysis through the transition integration carrier to obtain data priority support coefficients corresponding to each transition priority information.
For example, the priority of the transition level is determined based on the second target internet data backup network, in other words, the transition level data carrier is mined according to the depth characteristic of the transition level, then the parent data carrier and the transition level data carrier are subjected to carrier integration processing according to the carrier integration processing network of the transition level, so as to obtain the transition integration carrier, and then the transition level priority analysis is performed by the classification recognition network based on the transition level according to the transition integration carrier, so as to obtain the data priority support coefficient corresponding to each transition level priority information.
S40: and carrying out carrier integration processing on the father data carrier, the transition integration carrier and the son data carrier through a second target Internet data backup network to obtain a target son integration carrier, and calling the target son integration carrier to carry out son priority analysis to obtain target data priority support coefficients corresponding to the son priority information.
For example, the secondary priority is determined based on the second target internet data backup network, in other words, the secondary data carriers are mined based on the secondary depth feature mining network, then the carrier integration processing network is used for carrying out carrier integration processing on the parent data carrier and the transition level depth carrier summation result secondary data carrier to obtain the target secondary integration carrier, and then the secondary priority analysis is carried out by the target secondary integration carrier based on the classification recognition network of the secondary, so that the data priority support coefficient corresponding to each piece of secondary priority information is obtained.
S50: and obtaining target data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father-level priority information, the data priority support coefficient corresponding to each transition-level priority information and the target data priority support coefficient corresponding to each son-level priority information, wherein the target data priority comprises target father-level priority information, target transition-level priority information corresponding to the target father-level priority information and target son-level priority information corresponding to the target transition-level priority information.
For example, the data priority support coefficient corresponding to each father-level priority information, the data priority support coefficient corresponding to each transition-level priority information and the target data priority support coefficient corresponding to each son-level priority information can be obtained through the second target internet data backup network, the data priority support coefficient of each grading branch line is obtained, the father-level priority information, the transition-level priority information and the son-level priority information in the grading branch line corresponding to the maximum data priority support coefficient are screened out, and the data priority of the spare internet data output by the second target internet data backup network, namely the target father-level priority information, the target transition-level priority information corresponding to the target father-level priority information and the target son-level priority information corresponding to the target transition-level priority information, can enable the efficiency and the accuracy of priority determination to be higher according to the second target internet data backup network.
As an implementation manner, the debugging process of the internet data backup network provided in the embodiment of the present application may include the following steps:
s901: the method comprises the steps of obtaining an Internet data sample and a corresponding Internet data priority indication, loading the Internet data sample into a first basic Internet data backup network, and mining a priority description carrier of the Internet data sample through the first basic Internet data backup network to obtain a sample data priority description carrier.
The internet data sample is internet data adopted in the debugging process, the internet data priority indication is indication information containing hierarchical data priority and can be realized in a marked mode, the internet data priority indication comprises father-level priority information indication and corresponding son-level priority information indication, the first basic internet data backup network is a first internet data backup network after the initialization of network parameters, the first internet data backup network is an internet data backup network comprising father-level priority analysis and son-level priority analysis, the sample data priority description carrier is a priority description carrier corresponding to the internet data sample, and the priority characteristics of the sample data priority description carrier are characterized.
S902: and excavating the sample data priority description carrier through the first basic internet data backup network to obtain a sample parent data carrier, and analyzing the parent priority through the sample parent data carrier to obtain sample data priority support coefficients corresponding to each piece of parent priority information.
The sample father data carrier is the father data carrier obtained in the debugging process, the sample data priority support coefficient is probability information of the corresponding priority obtained in the debugging process as the corresponding priority of the internet data sample. For example, the priority of the father level is determined based on a father level priority analysis module in the first basic internet data backup network, the father level priority analysis module comprises a depth feature mining module and a classification mapping module, and the priority is determined through the depth feature mining module and the classification mapping module, so that sample data priority support coefficients corresponding to the priority information of each father level are obtained.
S903: the method comprises the steps of carrying out sub-level data carrier mining on a sample data priority description carrier through a first basic internet data backup network to obtain a sample sub-level data carrier, carrying out carrier integration processing on a sample parent data carrier and the sample sub-level data carrier to obtain a sample sub-level integration carrier, and carrying out sub-level priority analysis on the sample sub-level integration carrier to obtain sample data priority support coefficients corresponding to each piece of sub-level priority information, wherein a priority refinement index corresponding to the sub-level priority information is larger than a priority refinement index corresponding to the parent priority information.
The sample sub-level data carrier is a sub-level data carrier obtained in the debugging process, and the sample sub-level integration carrier is an integration carrier obtained in the debugging process of the first basic Internet data backup network. For example, the sub-level priority analysis module in the first basic internet data backup network is used for determining the sub-level priority, the sub-level priority analysis module comprises a depth feature mining module, a carrier integration processing module and a classification mapping module, and the sub-level priority analysis module is used for determining the priority based on the depth feature mining module, the carrier integration processing module and the classification mapping module to obtain the sample data priority support coefficient corresponding to each sub-level priority information.
S904: obtaining errors between sample data priority support coefficients corresponding to the priority information of each father level and father level priority information indications in the internet data priority indications to obtain father level priority information error information, and obtaining errors between sample data priority support coefficients corresponding to the priority information of each son level and the son level priority information indications in the internet data priority indications to obtain son level priority information error information.
The parent priority information error information represents an error between a sample data priority support coefficient corresponding to parent priority information obtained by debugging and a parent priority information indication in the internet data priority indication, and the child priority information error information represents an error between a sample data priority support coefficient corresponding to child priority obtained by debugging and a child priority information indication in the internet data priority indication. For example, based on a classification error algorithm (cross entropy, log likelihood, etc.), the error between the sample data priority support coefficient corresponding to each parent priority information and the parent priority information indication in the internet data priority indication is obtained, the parent priority information error information is obtained, and the error between the sample data priority support coefficient corresponding to each child priority information and the child priority information indication in the internet data priority indication is obtained, so as to obtain the child priority information error information.
S905: optimizing the first basic Internet data backup network through the parent priority information error information and the child priority information error information, and simultaneously returning to the step of obtaining the Internet data sample and the corresponding Internet data priority indication to repeat until the preset convergence requirement is met, so as to obtain the first target Internet data backup network.
The preset convergence requirement is, for example, that the debugging times reach the preset times, that the parameter of the network is not changed any more, and the like, and is not specifically limited, in other words, the network reaches convergence. The first target internet data backup network is a network for information multi-level prioritization after debugging. For example, the error sum of the parent priority information error information and the child priority information error information is obtained, then judgment is carried out, the first basic internet data backup network is optimized through error sum back propagation, the optimized (debugged) first basic internet data backup network is obtained, the optimized first basic internet data backup network is determined to be the first basic internet data backup network, and meanwhile, the step of obtaining the internet data sample and the corresponding internet data priority indication is returned to be repeated until the preset convergence requirement is met, and the first target internet data backup network is obtained.
Through the debugging process of the internet data backup network, the internet data sample is loaded into the first basic internet data backup network to carry out priority description carrier mining on the internet data sample, so as to obtain a sample data priority description carrier, the sample data priority description carrier is subjected to father data carrier mining, so as to obtain a sample father data carrier, father priority analysis is carried out on the sample father data carrier, so as to obtain sample data priority support coefficients corresponding to each father priority information, son data carrier mining is carried out on the sample data priority description carrier, so as to obtain a sample son data carrier, carrier integration processing is carried out on the sample father data carrier and the sample son data carrier, so as to obtain sample data priority support coefficients corresponding to each son priority information, and son priority analysis is carried out on the sample son integration carrier. Based on the above, the sample parent-level data carrier and the sample child-level data carrier are subjected to carrier integration processing to obtain the sample child-level integrated carrier, and then child-level priority analysis is performed, wherein the priority refinement index corresponding to the child-level priority information is larger than the priority refinement index corresponding to the parent-level priority information, so that the network can learn the hierarchical knowledge of the priority during debugging, the parent-level priority information error information and the child-level priority information error information are obtained, and repeated adjustment is performed through the parent-level priority information error information and the child-level priority information error information until the preset convergence requirement is met, so that the accuracy of the obtained first target Internet data backup network is improved.
As an embodiment, the debugging process of the internet data backup network may further include:
s906: and loading the internet data sample into a second basic internet data backup network, and mining the priority description carrier of the internet data sample through the second basic internet data backup network to obtain a sample data priority description carrier.
The second basic internet data backup network is a second internet data backup network after the initialization of the network parameter values is completed, and the second internet data backup network is an internet data backup network comprising father-level priority analysis, transition-level priority analysis and child-level priority analysis. For example, the internet data sample is loaded into the second basic internet data backup network, and the priority description carrier mining module is used for mining the internet data sample based on the priority carrier mining module in the second basic internet data backup network to obtain the sample data priority description carrier, wherein the priority carrier mining module can also be a quantization module which is debugged in advance.
S907: and excavating the sample data priority description carrier through the second basic Internet data backup network to obtain a sample parent data carrier, and analyzing the parent priority through the sample parent data carrier to obtain sample data priority support coefficients corresponding to each piece of parent priority information.
S908: and excavating the transition level data carrier by the sample data priority description carrier through the second basic internet data backup network to obtain a sample transition level data carrier, carrying out carrier integration processing by the sample parent data carrier and the sample transition level data carrier to obtain a sample transition integration carrier, and carrying out transition level priority analysis by the sample transition integration carrier to obtain sample data priority support coefficients corresponding to each transition level priority information.
The sample transition data carrier is a transition data carrier obtained in the process of debugging the second basic internet data backup network, and the sample transition integration carrier is an integration carrier obtained in the process of debugging the transition. The priority determination of the transition priority information is made, for example, based on a transition priority resolution network in the second underlying internet data backup network. The transition priority analysis network comprises a depth feature mining module, a carrier integration processing module and a classification mapping module, and the determination of the data priority of the transition level is carried out based on the depth feature mining module, the carrier integration processing module and the classification mapping module which are included in the transition priority analysis network, so that sample data priority support coefficients corresponding to the priority information of each transition level are obtained. The vector integration processing module is, for example, an integrated vector obtained by loading a vector for tail-end and gradient optimizing vector mining according to ResNet.
S909: and carrying out carrier integration treatment on the sample father-stage data carrier, the sample transition integration carrier and the sample son-stage data carrier through a second basic internet data backup network to obtain a sample target son-stage integration carrier, and calling the sample target son-stage integration carrier to carry out son-stage priority analysis to obtain a sample target data priority support coefficient corresponding to each piece of son-stage priority information, wherein the priority refinement index corresponding to the son-stage priority information is larger than the priority refinement index corresponding to the transition-stage priority information, and the priority refinement index corresponding to the transition-stage priority information is larger than the priority refinement index corresponding to the father-stage priority information.
The sample target sub-level integrating carrier is a target sub-level integrating carrier obtained in the process of debugging the second basic internet data backup network. The priority determination of the sub-level priority information is performed, for example, based on a sub-level priority resolution module in the second underlying internet data backup network. The sub-level priority analysis module comprises a depth feature mining module, a carrier integration processing module and a classification mapping module, and the sub-level priority analysis module is used for determining the data priority of the sub-level based on the depth feature mining module, the carrier integration processing module and the classification mapping module, so as to obtain sample data priority support coefficients corresponding to the priority information of each sub-level. The vector integration processing module is, for example, an integrated vector obtained by carrying out gradient optimization vector mining based on ResNet after loading the vector into the vector.
S910: obtaining errors between sample data priority support coefficients corresponding to the priority information of each father level and father level priority information indication in the internet data priority indication, obtaining father level priority information error information, obtaining errors between sample data priority support coefficients corresponding to the priority information of each transition level and transition level priority information indication in the internet data priority indication, obtaining transition level priority information error information, and obtaining errors between sample target data priority support coefficients corresponding to the priority information of each son level and the priority information indication of the internet data priority indication, thereby obtaining son level priority information target error information. The internet data priority indication comprises an indication corresponding to the transition priority. The transition priority information error information represents an error between a sample data priority support coefficient corresponding to the transition priority information obtained by debugging and a transition priority information indication in the internet data priority indication. For example, parent priority information error information, transition priority information error information, and child priority information error information are obtained based on a classification error algorithm (cross entropy, log likelihood, etc.).
S911: optimizing the second basic internet data backup network through the parent priority information error information, the transition priority information error information and the child priority information target error information, and simultaneously returning to the step of obtaining the internet data sample and the corresponding internet data priority indication for repeating until the preset convergence requirement is met, so as to obtain the second target internet data backup network.
For example, the summation result of the parent priority information error information, the transition priority information error information and the target error information of the child priority information is obtained, whether the cutoff requirement is met or not is evaluated through the summation result, if yes, the second basic internet data backup network is optimized based on the error information summation result, the optimized second basic internet data backup network is obtained, the optimized second basic internet data backup network is determined to be the second basic internet data backup network, and meanwhile, the steps of obtaining the internet data sample and the corresponding internet data priority indication are returned to be repeated until the preset convergence requirement is met, and the second target internet data backup network is obtained.
As an embodiment, the second internet data backup network may also be an internet data backup network comprising a parent priority resolution, at least two transition priority resolutions, and a child priority resolution. In the process of determining the priority of the internet data, the second target internet data backup network obtained through debugging can perform father-level priority analysis, multiple transition-level priority analysis and child-level priority analysis to obtain target father-level priority information, target transition-level priority information of multiple transition-level priority analysis corresponding to the target father-level priority information and target child-level priority information corresponding to the last target transition-level priority information. Based on the above process, based on obtaining the parent priority information error information, the transition priority information error information and the child priority information error information, optimizing the second basic internet data backup network based on the parent priority information error information, the transition priority information error information and the child priority information error information, and repeating until the second basic internet data backup network meets the preset convergence requirement, so as to obtain the second target internet data backup network, thereby enabling the accuracy of the second target internet data backup network to be higher.
For example, the embodiment of the present application further provides a complete internet data backup method, which specifically may include:
s1000: and acquiring the internet data to be backed up, loading the internet data to be backed up into a second target internet data backup network, and mining the priority description carrier of the internet data to be backed up through a priority description carrier mining module in the second target internet data backup network to obtain a data priority description carrier.
S2000: and excavating the data priority description carrier by a father-level priority analysis module in the second target Internet data backup network to obtain a father-level data carrier, and analyzing the father-level priority by the father-level data carrier to obtain data priority support coefficients corresponding to the priority information of each father-level.
S3000: and mining the data priority description carrier through an intermediate classification network in the second target Internet data backup network to obtain a transition data carrier, carrying out carrier integration processing through a father data carrier and the transition data carrier to obtain a transition integration carrier, and carrying out transition priority analysis through the transition integration carrier to obtain data priority support coefficients corresponding to each transition priority information.
S4000: and carrying out sub-level data carrier mining on the data priority description carrier through a sub-level priority analysis module in the second target Internet data backup network to obtain a sub-level data carrier, carrying out carrier integration processing on the father data carrier, the transition integration carrier and the sub-level data carrier to obtain a target sub-level integration carrier, and calling the target sub-level integration carrier to carry out sub-level priority analysis to obtain target data priority support coefficients corresponding to all sub-level priority information.
S5000: the method comprises the steps of obtaining data priority level information, screening out data priority support coefficients corresponding to current parent priority information from data priority support coefficients corresponding to each parent priority information, obtaining current transition priority information corresponding to the current parent priority information through the data priority level information, and obtaining current child priority information corresponding to the current transition priority information.
S6000: and determining the data priority support coefficient corresponding to the current transition priority information from the data priority support coefficients corresponding to the transition priority information, and determining the data priority support coefficient corresponding to the current sub-priority information from the target data priority support coefficient corresponding to the sub-priority information. And obtaining a multiplication result of the data priority support coefficient corresponding to the current parent priority information, the data priority support coefficient corresponding to the current transition priority information and the data priority support coefficient corresponding to the current child priority information to obtain the current level combination support coefficient.
S7000: traversing the data priority support coefficients corresponding to the priority information of each father level, the data priority support coefficients corresponding to the priority information of each transition level and the data priority support coefficients corresponding to the priority information of each son level to obtain each level combination support coefficient. Determining target level combination support coefficients from the level combination support coefficients, and determining target data priority corresponding to the internet data to be backed up through the target level combination support coefficients, wherein the target data priority comprises target father-level priority information, target transition-level priority information corresponding to the target father-level priority information and target son-level priority information corresponding to the target transition-level priority information.
According to the Internet data backup method, the server and the medium, internet data samples are loaded into a first basic Internet data backup network to conduct priority description carrier mining on the Internet data samples, so that sample data priority description carriers are obtained, father data carrier mining is conducted on the sample data priority description carriers, sample father data carriers are obtained, father data priority analysis is conducted on the sample father data carriers, sample data priority support coefficients corresponding to each father data priority information are obtained, son data carrier mining is conducted on the sample data priority description carriers, sample son data carriers are obtained, carrier integration processing is conducted on the sample father data carriers and the sample son data carriers, and sample son integration carriers are called to conduct son priority analysis, so that sample data priority support coefficients corresponding to each son priority information are obtained. In the network debugging process, carrier integration processing is carried out based on a sample father-stage data carrier and a sample son-stage data carrier to obtain a sample son-stage integration carrier, then son-stage priority analysis is carried out, wherein a priority refinement index corresponding to son-stage priority information is larger than a priority refinement index corresponding to father-stage priority information, based on the priority refinement index, the level composition of priority can be learned in the network debugging process, father-stage priority information error information and son-stage priority information error information can be obtained, optimized debugging is carried out based on the father-stage priority information error information and the son-stage priority information error information until a preset convergence requirement is met, and a first target Internet data backup network is obtained, so that the precision of the first target Internet data backup network is improved.
It should be noted that although the steps of the methods in the present application are depicted in the accompanying drawings in a particular order, this does not require or imply that the steps must be performed in that particular order, or that all illustrated steps be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform, etc.
The following describes an embodiment of an apparatus of the present application, which may be used to perform the method for backing up internet data in the foregoing embodiment of the present application. Fig. 2 schematically shows a block diagram of a data backup apparatus according to an embodiment of the present application. As shown in fig. 2, the data backup apparatus 200 includes:
the data acquisition module 210 is configured to acquire internet data to be backed up, and perform priority description carrier mining on the internet data to be backed up to obtain a data priority description carrier;
the carrier mining module 220 is configured to mine the data priority description carrier for a parent data carrier to obtain a parent data carrier, and perform parent priority analysis through the parent data carrier to obtain data priority support coefficients corresponding to each piece of parent priority information;
The carrier integration module 230 is configured to perform carrier mining on the data priority description carrier to obtain a sub-level data carrier, perform carrier integration processing on the parent data carrier and the sub-level data carrier to obtain a sub-level integrated carrier, and invoke the sub-level integrated carrier to perform sub-level priority analysis to obtain a data priority support coefficient corresponding to each piece of sub-level priority information, where a priority refinement index corresponding to the piece of sub-level priority information is greater than a priority refinement index corresponding to the parent priority information;
the priority determining module 240 is configured to obtain a data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to the priority information of each parent and the data priority support coefficient corresponding to the priority information of each child, where the data priority includes target parent priority information and target child priority information corresponding to the priority information of the target parent;
and the data backup module 250 is configured to backup the to-be-backed-up internet data based on the data priority corresponding to the to-be-backed-up internet data.
Specific details of the data backup device provided in each embodiment of the present application have been described in the corresponding method embodiments, and are not described herein.
FIG. 3 schematically illustrates a block diagram of a computer system for implementing a data backup server according to an embodiment of the present application.
It should be noted that, the computer system 300 of the data backup server shown in fig. 3 is only an example, and should not impose any limitation on the functions and the scope of use of the embodiments of the present application.
As shown in fig. 3, the computer system 300 includes a central processing unit 301 (Central Processing Unit, CPU) that can perform various appropriate actions and processes according to a program stored in a Read-Only Memory 302 (ROM) or a program loaded from a storage section 308 into a random access Memory 303 (Random Access Memory, RAM). In the random access memory 303, various programs and data required for the system operation are also stored. The central processing unit 301, the read only memory 302, and the random access memory 303 are connected to each other via a bus 304. An Input/Output interface 305 (i.e., an I/O interface) is also connected to bus 304.
The following components are connected to the input/output interface 305: an input section 306 including a keyboard, a mouse, and the like; an output portion 307 including a Cathode Ray Tube (CRT), a liquid crystal display (Liquid Crystal Display, LCD), and the like, a speaker, and the like; a storage section 308 including a hard disk or the like; and a communication section 309 including a network interface card such as a local area network card, a modem, or the like. The communication section 309 performs communication processing via a network such as the internet. The driver 310 is also connected to the input/output interface 305 as needed. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed on the drive 310 as needed, so that a computer program read therefrom is installed into the storage section 308 as needed.
In particular, according to embodiments of the present application, the processes described in the various method flowcharts may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via the communication portion 309, and/or installed from the removable medium 311. The computer program, when executed by the central processor 301, performs the various functions defined in the system of the present application.
It should be noted that, the computer readable medium shown in the embodiments of the present application may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-Only Memory (ROM), an erasable programmable read-Only Memory (Erasable Programmable Read Only Memory, EPROM), flash Memory, an optical fiber, a portable compact disc read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present application, however, a computer-readable signal medium may include a data signal that propagates in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functions of two or more modules or units described above may be embodied in one module or unit, in accordance with embodiments of the present application. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.
From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a usb disk, a mobile hard disk, etc.) or on a network, and includes several instructions to cause a computing device (may be a personal computer, a server, a touch terminal, or a network device, etc.) to perform the method according to the embodiments of the present application.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains.
It is to be understood that the present application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (10)

1. An internet data backup method, applied to a data backup server, comprising:
acquiring internet data to be backed up, and mining a priority description carrier of the internet data to be backed up to obtain a data priority description carrier;
the data priority description carrier is subjected to father data carrier mining to obtain father data carriers, and father priority analysis is carried out through the father data carriers to obtain data priority support coefficients corresponding to the father priority information;
carrying out sub-level data carrier mining on the data priority description carrier to obtain a sub-level data carrier, carrying out carrier integration processing on the parent data carrier and the sub-level data carrier to obtain a sub-level integrated carrier, and calling the sub-level integrated carrier to carry out sub-level priority analysis to obtain a data priority support coefficient corresponding to each piece of sub-level priority information, wherein the priority refinement index corresponding to the sub-level priority information is larger than the priority refinement index corresponding to the parent priority information;
obtaining the data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father priority information and the data priority support coefficient corresponding to each son priority information, wherein the data priority comprises target father priority information and target son priority information corresponding to the target father priority information;
And backing up the internet data to be backed up based on the data priority corresponding to the internet data to be backed up.
2. The method according to claim 1, wherein said carrier integration processing by said parent data carrier and said child data carrier results in a child-level integrated carrier, comprising:
the father-stage data carrier and the son-stage data carrier are subjected to tail-end connection to obtain a tail-end data carrier;
performing gradient optimization carrier mining through the armature data carrier and the sub-level data carrier to obtain a gradient optimization carrier, and performing normalization treatment on the gradient optimization carrier to obtain the sub-level integration carrier;
wherein the gradient optimization carrier mining is performed through the armature data carrier and the sub-level data carrier to obtain a gradient optimization carrier, and the gradient optimization carrier comprises the following steps:
and carrying out embedded coding on the armature data carrier according to a preset gradient optimization variable to obtain an embedded coding carrier, and obtaining a carrier summation result of the embedded coding carrier and the sub-level data carrier to obtain the gradient optimization carrier.
3. The method according to claim 1, wherein the obtaining the data priority corresponding to the to-be-backed-up internet data by using the data priority support coefficient corresponding to the respective parent priority information and the data priority support coefficient corresponding to the respective child priority information includes:
Acquiring data priority level information;
integrating the data priority support coefficients corresponding to the parent priority information and the data priority support coefficients corresponding to the child priority information through the data priority level information to obtain each level combination support coefficient;
determining a target level combination support coefficient in the level combination support coefficients, and determining a data priority corresponding to the internet data to be backed up through the target level combination support coefficient;
the integrating, by the data priority level information, the data priority support coefficient corresponding to the parent level priority information and the data priority support coefficient corresponding to the child level priority information to obtain each level combination support coefficient includes:
screening out the data priority support coefficient corresponding to the current parent priority information from the data priority support coefficients corresponding to the parent priority information;
acquiring current sub-level priority information corresponding to the current parent level priority information through the data priority level information, and determining a data priority support coefficient corresponding to the current sub-level priority information from the data priority support coefficients corresponding to the sub-level priority information;
Obtaining a multiplication result of the data priority support coefficient corresponding to the current parent priority information and the data priority support coefficient corresponding to the current child priority information to obtain a current level combination support coefficient;
and traversing the data priority support coefficients corresponding to the priority information of each father level and the data priority support coefficients corresponding to the priority information of each son level to obtain the combined support coefficients of each level.
4. The method according to claim 1, wherein the method further comprises:
loading the internet data to be backed up into a first target internet data backup network;
the first target internet data backup network is used for carrying out priority description carrier mining on the internet data to be backed up to obtain a data priority description carrier;
the data priority description carrier is subjected to father data carrier mining through the first target Internet data backup network to obtain father data carriers, father priority analysis is carried out through the father data carriers, and data priority support coefficients corresponding to the father priority information are obtained;
the data priority description carrier is subjected to sub-level data carrier mining through the first target Internet data backup network to obtain a sub-level data carrier, carrier integration processing is carried out through the father-level data carrier and the sub-level data carrier to obtain a sub-level integration carrier, and the sub-level integration carrier is called to carry out sub-level priority analysis to obtain data priority support coefficients corresponding to each piece of sub-level priority information;
The first target Internet data backup network comprises a priority description carrier mining module, a father priority analysis module and a son priority analysis module; the loading the internet data to be backed up into the first target internet data backup network includes:
loading the internet data to be backed up into the priority description carrier mining module to mine the priority description carrier, so as to obtain a data priority description carrier;
loading the data priority description carrier into the father priority analysis module, mining the data priority description carrier by the father priority analysis module to obtain a father data carrier, and obtaining data priority support coefficients corresponding to the father priority information by the father data carrier;
loading the data priority description carrier and the father stage data carrier into the child priority analysis module, carrying out child stage data carrier mining on the data priority description carrier through the child priority analysis module to obtain a child stage data carrier, carrying out carrier integration processing on the father stage data carrier and the child stage data carrier to obtain a child stage integration carrier, and calling the child stage integration carrier to obtain a child stage priority information support coefficient to obtain a data priority support coefficient corresponding to each piece of child stage priority information;
The sub-level priority analyzing module comprises a sub-level data carrier mining sub-module, a carrier integration processing sub-module and a sub-level priority analyzing sub-module; the loading the data priority description carrier and the parent data carrier into the child priority parsing module includes:
loading the data priority description carrier into the sub-level data carrier mining submodule to carry out sub-level data carrier mining to obtain the sub-level data carrier;
loading the parent data carrier and the child data carrier into the carrier integration processing sub-module for carrier integration processing to obtain a child integration carrier;
and loading the sub-level integrating carrier into the sub-level priority analysis sub-module to acquire sub-level priority information support coefficients, so as to acquire the data priority support coefficients corresponding to the sub-level priority information.
5. The method according to claim 1, wherein the method further comprises:
mining the data priority description carrier into a transition data carrier to obtain the transition data carrier;
carrying out carrier integration treatment through the father data carrier and the transition level data carrier to obtain a transition integration carrier, and carrying out transition level priority analysis through the transition integration carrier to obtain data priority support coefficients corresponding to each transition level priority information;
Carrying out carrier integration treatment on the father-stage data carrier, the transition integration carrier and the son-stage data carrier to obtain a target son-stage integration carrier, and calling the target son-stage integration carrier to carry out son-stage priority analysis to obtain a target data priority support coefficient corresponding to each piece of son-stage priority information, wherein the priority refinement index corresponding to the son-stage priority information is larger than the priority refinement index corresponding to the transition-stage priority information, and the priority refinement index corresponding to the transition-stage priority information is larger than the priority refinement index corresponding to the father-stage priority information;
obtaining a target data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father priority information, the data priority support coefficient corresponding to each transition priority information and the target data priority support coefficient corresponding to each son priority information, wherein the target data priority comprises target father priority information, target transition priority information corresponding to the target father priority information and target son priority information corresponding to the target transition priority information;
Wherein, the carrier integration treatment is carried out by the father-stage data carrier and the transition-stage data carrier to obtain a transition integration carrier, which comprises the following steps: the father-stage data carrier and the transition-stage data carrier are subjected to tail-joining to obtain a transition tail-joining data carrier; performing gradient optimization carrier mining through the transition armature data carrier and the sub-level data carrier to obtain a transition gradient optimization carrier; normalizing the transition gradient optimization vector to obtain the transition integration vector;
the carrier integration treatment is carried out on the father stage data carrier, the transition integration carrier and the son stage data carrier to obtain a target son stage integration carrier, which comprises the following steps: performing tail-joining on the father stage data carrier, the transition integration carrier and the son stage data carrier to obtain a target tail-joining data carrier; performing gradient optimization carrier mining through the target armature data carrier and the sub-level data carrier to obtain a target gradient optimization carrier; and normalizing the target gradient optimization vector to obtain the target sub-level integration vector.
6. The method according to claim 5, further comprising:
Loading the internet data to be backed up into a second target internet data backup network;
the second target internet data backup network is used for carrying out priority description carrier mining on the internet data to be backed up to obtain a data priority description carrier;
the data priority description carrier is subjected to father data carrier mining through the second target Internet data backup network to obtain father data carriers, father priority analysis is carried out through the father data carriers, and data priority support coefficients corresponding to the father priority information are obtained;
mining the data priority description carrier through the second target Internet data backup network to obtain a transition data carrier, carrying out carrier integration processing through the father data carrier and the transition data carrier to obtain a transition integration carrier, and carrying out transition priority analysis through the transition integration carrier to obtain data priority support coefficients corresponding to each transition priority information;
carrying out carrier integration processing on the father data carrier, the transition integration carrier and the son data carrier through the second target Internet data backup network to obtain a target son integration carrier, and calling the target son integration carrier to carry out son priority analysis to obtain target data priority support coefficients corresponding to each piece of son priority information;
Obtaining the target data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each father priority information, the data priority support coefficient corresponding to each transition priority information and the target data priority support coefficient corresponding to each son priority information, wherein the target data priority comprises target father priority information, target transition priority information corresponding to the target father priority information and target son priority information corresponding to the target transition priority information.
7. The method according to any one of claims 4 to 6, further comprising a training process of an internet data backup network, comprising the steps of:
acquiring an Internet data sample and a corresponding Internet data priority indication;
loading the internet data sample into a first basic internet data backup network, and mining a priority description carrier of the internet data sample through the first basic internet data backup network to obtain a sample data priority description carrier;
the sample data priority description carrier is subjected to father data carrier mining through the first basic internet data backup network to obtain a sample father data carrier, and father priority analysis is performed through the sample father data carrier to obtain sample data priority support coefficients corresponding to each piece of father priority information;
Carrying out sub-level data carrier mining on the sample data priority description carrier through the first basic internet data backup network to obtain a sample sub-level data carrier, carrying out carrier integration processing on the sample parent level data carrier and the sample sub-level data carrier to obtain a sample sub-level integration carrier, and calling the sample sub-level integration carrier to carry out sub-level priority analysis to obtain a sample data priority support coefficient corresponding to each piece of sub-level priority information, wherein the priority refinement index corresponding to the sub-level priority information is larger than the priority refinement index corresponding to the parent level priority information;
obtaining errors between sample data priority support coefficients corresponding to the priority information of each father level and father level priority information indications in the internet data priority indications to obtain father level priority information error information, and obtaining errors between sample data priority support coefficients corresponding to the priority information of each son level and the internet data priority indication to obtain son level priority information error information;
and optimizing the first basic Internet data backup network through the parent priority information error information and the child priority information error information, and returning to the step of acquiring the Internet data sample and the corresponding Internet data priority indication at the same time to repeat until the preset convergence requirement is met, so as to acquire the first target Internet data backup network.
8. The method of claim 7, wherein the method further comprises:
loading the internet data sample into a second basic internet data backup network, and mining a priority description carrier of the internet data sample through the second basic internet data backup network to obtain a sample data priority description carrier;
the sample data priority description carrier is subjected to father data carrier mining through the second basic internet data backup network to obtain a sample father data carrier, and father priority analysis is performed through the sample father data carrier to obtain sample data priority support coefficients corresponding to each piece of father priority information;
the sample data priority description carrier is subjected to transition level data carrier mining through the second basic internet data backup network to obtain a sample transition level data carrier, carrier integration processing is carried out through the sample father level data carrier and the sample transition level data carrier to obtain a sample transition integration carrier, transition level priority analysis is carried out through the sample transition integration carrier, and sample data priority support coefficients corresponding to each piece of transition level priority information are obtained;
Carrying out carrier integration treatment on the sample father stage data carrier, the sample transition integration carrier and the sample son stage data carrier through the second basic internet data backup network to obtain a sample target son stage integration carrier, and calling the sample target son stage integration carrier to carry out son stage priority analysis to obtain a sample target data priority support coefficient corresponding to each piece of son stage priority information, wherein the priority refinement index corresponding to the son stage priority information is larger than the priority refinement index corresponding to the transition stage priority information, and the priority refinement index corresponding to the transition stage priority information is larger than the priority refinement index corresponding to the father stage priority information;
obtaining errors between sample data priority support coefficients corresponding to the priority information of each father level and father level priority information indications in the internet data priority indications, obtaining father level priority information error information, obtaining errors between sample data priority support coefficients corresponding to the priority information of each transition level and transition level priority information indications in the internet data priority indications, obtaining transition level priority information error information, and obtaining errors between sample target data priority support coefficients corresponding to the priority information of each child level and the internet data priority indication neutron level priority information indications, thereby obtaining child level priority information target error information;
Optimizing the second basic internet data backup network through the parent priority information error information, the transition priority information error information and the child priority information target error information, and simultaneously returning to the step of obtaining the internet data sample and the corresponding internet data priority indication for repeating until the preset convergence requirement is met, so as to obtain the second target internet data backup network.
9. A data backup apparatus, comprising:
the data acquisition module is used for acquiring the internet data to be backed up, and carrying out priority description carrier mining on the internet data to be backed up to obtain a data priority description carrier;
the carrier mining module is used for mining the data priority description carrier into a father data carrier to obtain the father data carrier, and carrying out father priority analysis through the father data carrier to obtain data priority support coefficients corresponding to each piece of father priority information;
the carrier integration module is used for carrying out carrier mining on the data priority description carrier to obtain a child data carrier, carrying out carrier integration processing on the parent data carrier and the child data carrier to obtain a child level integration carrier, and calling the child level integration carrier to carry out child level priority analysis to obtain a data priority support coefficient corresponding to each piece of child level priority information, wherein the priority refinement index corresponding to the child level priority information is larger than the priority refinement index corresponding to the parent level priority information;
The priority determining module is configured to obtain a data priority corresponding to the internet data to be backed up according to the data priority support coefficient corresponding to each parent priority information and the data priority support coefficient corresponding to each child priority information, where the data priority includes target parent priority information and target child priority information corresponding to the target parent priority information;
and the data backup module is used for backing up the internet data to be backed up based on the data priority corresponding to the internet data to be backed up.
10. A data backup server, comprising:
a processor;
and a memory for storing executable instructions of the processor;
wherein the processor is configured to perform the method of any one of claims 1 to 8 via execution of the executable instructions.
CN202310364120.9A 2023-04-07 2023-04-07 Internet data backup method, device and server Active CN116107815B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310364120.9A CN116107815B (en) 2023-04-07 2023-04-07 Internet data backup method, device and server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310364120.9A CN116107815B (en) 2023-04-07 2023-04-07 Internet data backup method, device and server

Publications (2)

Publication Number Publication Date
CN116107815A true CN116107815A (en) 2023-05-12
CN116107815B CN116107815B (en) 2023-06-13

Family

ID=86267533

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310364120.9A Active CN116107815B (en) 2023-04-07 2023-04-07 Internet data backup method, device and server

Country Status (1)

Country Link
CN (1) CN116107815B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103227835A (en) * 2013-04-27 2013-07-31 苏州洁祥电子有限公司 Vehicle networking system and data backup method thereof
US20140081920A1 (en) * 2012-09-19 2014-03-20 Fujitsu Limited Medium, control method, and information processing apparatus
CN112612644A (en) * 2020-12-24 2021-04-06 深圳市科力锐科技有限公司 Host data backup method, device, storage medium and device
CN114647537A (en) * 2020-12-18 2022-06-21 伊姆西Ip控股有限责任公司 Method for backup and restore
CN115794494A (en) * 2022-12-27 2023-03-14 中国电信股份有限公司 Data backup method, system, device, equipment and medium based on dynamic strategy

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140081920A1 (en) * 2012-09-19 2014-03-20 Fujitsu Limited Medium, control method, and information processing apparatus
CN103227835A (en) * 2013-04-27 2013-07-31 苏州洁祥电子有限公司 Vehicle networking system and data backup method thereof
CN114647537A (en) * 2020-12-18 2022-06-21 伊姆西Ip控股有限责任公司 Method for backup and restore
CN112612644A (en) * 2020-12-24 2021-04-06 深圳市科力锐科技有限公司 Host data backup method, device, storage medium and device
CN115794494A (en) * 2022-12-27 2023-03-14 中国电信股份有限公司 Data backup method, system, device, equipment and medium based on dynamic strategy

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
《2022 IEEE 13TH INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING AND SERVICE SCIENCE (ICSESS)》: "A Cross-platform Errand Service Application for Campus with a Dual-user Identity Model", 《2022 IEEE 13TH INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING AND SERVICE SCIENCE (ICSESS)》, pages 155 - 160 *
杨宏胜: "基于大数据的用户个性化推荐***设计与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》, pages 138 - 261 *

Also Published As

Publication number Publication date
CN116107815B (en) 2023-06-13

Similar Documents

Publication Publication Date Title
US10296307B2 (en) Method and system for template extraction based on source code similarity
CN112163226B (en) Binary function similarity detection method based on graph automatic encoder
US20220207304A1 (en) Robustness setting device, robustness setting method, storage medium storing robustness setting program, robustness evaluation device, robustness evaluation method, storage medium storing robustness evaluation program, computation device, and storage medium storing program
CN114357170A (en) Model training method, analysis method, device, equipment and medium
CN115660859A (en) Full-link automatic testing method and system based on Internet finance
CN116107815B (en) Internet data backup method, device and server
CN112684396B (en) Data preprocessing method and system for electric energy meter operation error monitoring model
CN104598705A (en) Method and device for recognizing underground material layer
CN111325578B (en) Sample determination method and device of prediction model, medium and equipment
CN116861373A (en) Query selectivity estimation method, system, terminal equipment and storage medium
CN115098389B (en) REST interface test case generation method based on dependency model
Passos et al. Multi-objective optimization with Kriging surrogates using “moko”, an open source package
CN115587029A (en) Patch detection method and device, electronic equipment and computer readable medium
CN114780439A (en) Reuse method of test cases among similar programs facing to parameter path flow graph
EP3547154B1 (en) Constraint satisfaction software tool for database tables
CN117561502A (en) Method and device for determining failure reason
CN114429166A (en) Method, device and equipment for acquiring high-dimensional features of data and computer storage medium
US20230144390A1 (en) Non-transitory computer-readable storage medium for storing operation program, operation method, and calculator
CN115758135B (en) Track traffic signal system function demand tracing method and device and electronic equipment
CN112837148B (en) Risk logic relationship quantitative analysis method integrating domain knowledge
CN113806558B (en) Question selection method, knowledge graph construction device and electronic equipment
CN116451678B (en) Data relation recognition and data table integration method
CN117234737A (en) Cloud manufacturing scheduling simulation method and system based on micro-service architecture
CN110795941B (en) Named entity identification method and system based on external knowledge and electronic equipment
CN117609005A (en) Code similarity detection method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant