CN115330319B - Logistics information processing method based on big data and artificial intelligence server - Google Patents

Logistics information processing method based on big data and artificial intelligence server Download PDF

Info

Publication number
CN115330319B
CN115330319B CN202211250431.4A CN202211250431A CN115330319B CN 115330319 B CN115330319 B CN 115330319B CN 202211250431 A CN202211250431 A CN 202211250431A CN 115330319 B CN115330319 B CN 115330319B
Authority
CN
China
Prior art keywords
distribution
resource distribution
information
resource
examples
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.)
Active
Application number
CN202211250431.4A
Other languages
Chinese (zh)
Other versions
CN115330319A (en
Inventor
邹春风
高春亚
张菊
杨建国
陈佳妮
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.)
Chuangyu Intelligent Changshu Netlink Technology Co ltd
Original Assignee
Chuangyu Intelligent Changshu Netlink Technology 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 Chuangyu Intelligent Changshu Netlink Technology Co ltd filed Critical Chuangyu Intelligent Changshu Netlink Technology Co ltd
Priority to CN202211250431.4A priority Critical patent/CN115330319B/en
Publication of CN115330319A publication Critical patent/CN115330319A/en
Application granted granted Critical
Publication of CN115330319B publication Critical patent/CN115330319B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Quality & Reliability (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application provides a logistics information processing method based on big data and an artificial intelligence server; enabling an artificial intelligence neural network according to a plurality of first distribution resource distribution information examples of the first intelligent logistics distribution environment to obtain a plurality of second distribution resource distribution information examples; training a first resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second intelligent logistics distribution environment and corresponding type differentiation keywords respectively; selecting information according to the quantized trusted data for a plurality of second distribution resource distribution information examples; and training a second resource distribution differentiation analysis network according to the obtained second distribution resource distribution information example. Through the method and the device, the distribution resource distribution information examples of different intelligent logistics distribution environments can be intelligently obtained, and the accuracy and timeliness of the differential analysis of the distribution resource distribution information are further improved.

Description

Logistics information processing method based on big data and artificial intelligence server
Technical Field
The application relates to the technical field of artificial intelligence and information processing, in particular to a logistics information processing method based on big data and an artificial intelligence server.
Background
Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, techniques and applications for simulating, extending and expanding human intelligence. Artificial Intelligence (AI) is often described by different terms, such as: machine learning, deep learning, natural language processing, predictive analysis, and the like. All the above points out that in the future, our platform and system are clever enough to learn our interaction and data, which will not only help us answer questions, but also predict our needs, manage our tasks and remind in time. Artificial Intelligence (AI) can connect many points in our lives (home, work, travel), bringing us a seamless experience from home to car to office. Most of the experience comes from our cell phone. The traditional logistics industry is also performing intelligent transformation thanks to the development of artificial intelligence, however, the rapid and rapidly increased logistics demand makes the current logistics resource scheduling technology difficult to provide a rapid and accurate response scheme.
Disclosure of Invention
In order to solve the technical problems in the related art, the application provides a logistics information processing method based on big data and an artificial intelligence server.
The application provides a logistics information processing method based on big data, which is characterized in that the logistics information processing method is applied to an artificial intelligence server and comprises the following steps:
enabling an artificial intelligence neural network according to a plurality of first distribution resource distribution information examples of a first intelligent logistics distribution environment to obtain a plurality of second distribution resource distribution information examples which are matched with the plurality of first distribution resource distribution information examples in a one-to-one mode; wherein the plurality of second distribution resource distribution information examples employ a second intelligent logistics distribution environment different from the first intelligent logistics distribution environment;
training a first resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second intelligent logistics distribution environment and corresponding type differentiation keywords respectively;
selecting and processing the plurality of second distribution resource distribution information examples according to information of quantized credible data through the trained first resource distribution differentiated analysis network;
training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to a second distribution resource distribution information example obtained by the information selection processing; wherein the architecture complexity of the second resource distribution differentiated analysis network is greater than the architecture complexity of the first resource distribution differentiated analysis network.
An independently implementable embodiment, before training a second resource distribution differentiation analysis network for the second intelligent logistics distribution environment based on a second distribution resource distribution information instance obtained by the information selection process, the method further comprises:
determining the number of target examples matched with the network performance quantitative data which can be used for training the second resource distribution differentiated analysis network according to a mapping list of the network performance quantitative data of the resource distribution differentiated analysis network and the number of the distribution resource distribution information examples which can be digested in a set time sequence step;
and selecting the distribution information examples corresponding to the target example number from a network training set consisting of the second distribution resource distribution information examples obtained by the information selection processing as examples for training a second resource distribution differentiated analysis network used for the second intelligent logistics distribution environment.
An independently implementable embodiment, the training of the first resource distribution differential analysis network for the second smart logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and respectively corresponding type-differential keywords, comprises:
performing an ith round of training on the first resource distribution differentiation analysis network according to a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and the type differentiation keywords respectively corresponding to the third distribution resource distribution information examples;
performing ith round information selection processing according to quantitative credible data on the plurality of second distribution resource distribution information examples through the first resource distribution differential analysis network trained in the ith round;
performing (i + 1) th round training on the first resource distribution differentiation analysis network according to the previous i round information selection result, the third distribution resource distribution information examples and the type differentiation keywords corresponding to the third distribution resource distribution information examples respectively;
taking the first resource distribution differential analysis network of the ith round of training as the first resource distribution differential analysis network after completing the training; wherein i is an integer with constraint conditions meeting the condition that i is more than or equal to 1 and less than or equal to j-1, and the value is added by one from 1, and j is an integer more than 2 and is used for representing the total cumulative round number of the cyclic training.
An independently implementable embodiment, the training of the first resource distribution differentiation analysis network for the second smart logistics distribution environment according to the plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and the respectively corresponding type differentiation keywords, comprises:
estimating a plurality of third distribution resource distribution information examples of the second intelligent logistics distribution environment through the first resource distribution differentiation analysis network to obtain quantitative credible data of estimation types corresponding to the plurality of third distribution resource distribution information examples respectively;
forming a cost function of the first resource distribution differentiation analysis network according to the quantitative credible data of the estimation type and the type differentiation keywords of the third distribution resource distribution information example;
updating parameters of the first resource distribution differential analysis network until the cost function meets a set condition, and taking the updated parameters of the first resource distribution differential analysis network when the cost function meets the set condition as the parameters of the trained first resource distribution differential analysis network;
correspondingly, the estimating, by the first resource distribution differentiation analysis network, the multiple third distribution resource distribution information examples of the second smart logistics distribution environment to obtain quantitative credible data of estimation types corresponding to the multiple third distribution resource distribution information examples, includes:
for any one of the plurality of third delivery resource distribution information examples, implementing the following steps:
implementing the following steps through the first resource distribution differentiation analysis network:
performing feature extraction processing on the third distribution resource distribution information example to obtain a visual key description of the third distribution resource distribution information example;
merging the visual key descriptions of the third distribution resource distribution information example to obtain a merged key description;
and performing non-limiting projection processing on the merged key description to obtain quantitative credible data of an estimation type corresponding to the third distribution resource distribution information example.
An independently implementable embodiment, the first resource distribution differential analysis network comprises a plurality of Sigmoid function units associated with each other; the performing non-limiting projection processing on the merged key description to obtain quantitative credible data of an estimation type corresponding to the third distribution resource distribution information example includes:
performing projection operation of a first Sigmoid function unit of the plurality of correlated Sigmoid function units on the merged key description through the first Sigmoid function unit;
transmitting the projection result of the first Sigmoid function unit to downstream Sigmoid function units which are mutually associated, and continuing to perform projection operation and projection result output in the downstream Sigmoid function units which are mutually associated until the projection result is transmitted to the Sigmoid function unit positioned at the tail of an associated path;
and taking the triggering result output by the Sigmoid function unit positioned at the end of the associated path as quantitative credible data of the estimation type corresponding to the third distribution resource distribution information example.
An independently implementable embodiment, said training a second resource distribution differential analysis network for said second intelligent logistics distribution environment according to a second distribution resource distribution information example obtained by said information selection processing, comprises:
determining thermodynamic diagrams of a plurality of types of second distribution resource distribution information examples obtained by the information selection processing;
when the second distribution resource distribution information example obtained by the information selection processing conforms to thermodynamic diagram coordination indexes in a plurality of types and the number of each type is greater than the corresponding type number threshold value, arbitrarily grabbing the distribution resource distribution information example corresponding to the type number threshold value from the distribution resource distribution information examples of each type in the second distribution resource distribution information example obtained by the information selection processing to form a network training set;
and training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the network training set.
An independently implementable embodiment, the training of the second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the second distribution resource distribution information example obtained by the information selection processing, comprises:
when the thermodynamic diagrams of the second distribution resource distribution information examples obtained by the information selection processing do not accord with the thermodynamic diagram coordination indexes in the multiple types, performing extended optimization according to the description of the neighbor resources aiming at the second distribution resource distribution information examples of the corresponding types so that the thermodynamic diagrams of the second distribution resource distribution information examples obtained by the extended optimization accord with the thermodynamic diagram coordination indexes in the multiple types;
forming a network training set according to the second distribution resource distribution information example obtained by the expansion optimization;
and training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the network training set.
An independently implementable embodiment, the training of the second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the second distribution resource distribution information example obtained by the information selection processing, comprises:
forming a network training set according to the plurality of third distribution resource distribution information examples and a second distribution resource distribution information example obtained by the information selection processing, and training a second resource distribution differentiation analysis network used for the second intelligent logistics distribution environment according to the network training set;
correspondingly, forming a network training set according to the plurality of third distribution resource distribution information examples and the second distribution resource distribution information examples obtained by the information selection processing, includes:
and traversing each type of the second distribution resource distribution information example obtained by the information selection processing, and implementing the following steps:
when the number of second distribution resource distribution information examples in the type is lower than a type number threshold of the type, adding a third distribution resource distribution information example of the type arbitrarily grabbed from the plurality of third distribution resource distribution information examples into the second distribution resource distribution information examples of the type to update the second distribution resource distribution information examples obtained by the information selection processing;
and forming a network training set according to a second distribution resource distribution information example obtained by the updated information selection processing.
An independently implementable embodiment, said performing, by the trained first resource distribution differential analysis network, information selection processing according to quantified credible data on the plurality of second distribution resource distribution information examples, comprises:
performing the following steps for any one of the plurality of second dispatch resource distribution information examples:
estimating the second distribution resource distribution information example through the trained first resource distribution differential analysis network to obtain quantitative credible data of multiple estimation types corresponding to the second distribution resource distribution information example;
determining the type difference keywords of the first distribution resource distribution information example corresponding to the second distribution resource distribution information example as the type difference keywords of the second distribution resource distribution information example;
and according to the quantitative credible data of a plurality of estimation types corresponding to the second distribution resource distribution information examples and the type differentiation keywords of the second distribution resource distribution information examples, taking the second distribution resource distribution information examples larger than the quantitative credible data threshold value as second distribution resource distribution information examples obtained by information selection processing.
The application also provides an artificial intelligence server, which comprises a processor and a memory; the processor is connected with the memory in communication, and the processor is used for reading the computer program from the memory and executing the computer program to realize the method.
The technical scheme provided by the embodiment of the application can have the following beneficial effects.
Acquiring a second distribution resource distribution information example of a second intelligent logistics distribution environment different from the first intelligent logistics distribution environment through an artificial intelligent neural network, and performing information selection on the second distribution resource distribution information example through a first resource distribution differentiation analysis network, so that the distribution resource distribution information examples of different intelligent logistics distribution environments are intelligently acquired, and distribution resource scheduling deviation caused by lack of the distribution resource distribution information examples is weakened; furthermore, a second resource distribution differentiation analysis network is trained through a distribution resource distribution information example with relatively high value quantity obtained through information selection, so that the second resource distribution differentiation analysis network can carry out accurate real-time resource distribution differentiation analysis on distribution resource distribution information, the accuracy and timeliness of the differentiation analysis on the distribution resource distribution information are further improved, various logistics distribution demands can be flexibly and accurately met through the second resource distribution differentiation analysis network, and the differentiation analysis result of the distribution resource distribution information is matched with the various logistics distribution demands to achieve rapid, flexible and accurate logistics resource scheduling.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic hardware structure diagram of an artificial intelligence server according to an embodiment of the present disclosure.
Fig. 2 is a schematic flow chart of a logistics information processing method based on big data according to an embodiment of the present application.
Fig. 3 is a block diagram of a logistics information processing device based on big data according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the application, as detailed in the appended claims.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
The method provided by the embodiment of the application can be executed in an artificial intelligence server, a computer device or a similar operation device. Taking an example of the artificial intelligence server running on the artificial intelligence server, fig. 1 is a hardware structure block diagram of the artificial intelligence server implementing the logistics information processing method based on big data according to the embodiment of the present application. As shown in FIG. 1, the artificial intelligence server 10 can include one or more (only one shown in FIG. 1) processors 102 (the processors 102 can include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, and optionally, the artificial intelligence server 10 can also include a transmission device 106 for communication functions. It will be understood by those skilled in the art that the configuration shown in FIG. 1 is merely illustrative and is not intended to limit the configuration of the artificial intelligence server 10 described above. For example, artificial intelligence server 10 may also include more or fewer components than are shown in FIG. 1, or have a different configuration than that shown in FIG. 1.
The memory 104 may be used to store computer programs, for example, software programs and modules of application software, such as a computer program corresponding to the logistics information processing method based on big data in the embodiment of the present application, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the methods described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located from the processor 102, which may be connected to the artificial intelligence server 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used for receiving or transmitting data via a network. The above-described specific example of the network may include a wireless network provided by a communication provider of the artificial intelligence server 10. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
Referring to fig. 2, fig. 2 is a schematic flow chart of a logistics information processing method based on big data according to an embodiment of the present application, which is applied to an artificial intelligence server, and the method is specifically described in detail by following technical solutions recorded in steps 210 to 240.
Step 210, enabling an artificial intelligence neural network according to a plurality of first distribution resource distribution information examples of the first intelligent logistics distribution environment to obtain a plurality of second distribution resource distribution information examples which are matched with the plurality of first distribution resource distribution information examples in a one-to-one mode.
In the practical implementation process, the first intelligent logistics distribution environment can be understood as different distribution states or different distribution processes. For example: logistics sorting environment information, logistics warehousing link information and logistics classification link information. It can also be understood as the regional information of the actual distribution area. The first dispatch resource distribution information example can be understood as a first dispatch resource distribution information sample. Further, the distribution information of the delivery resources may be understood as the configuration of different delivery resources, such as a manual distribution, a vehicle distribution, or a production line distribution. The artificial intelligence neural network may be a feedforward neural network based on artificial intelligence learning. Wherein the plurality of second distribution resource distribution information instances employ a second intelligent logistics distribution environment that is distinct from the first intelligent logistics distribution environment.
Illustratively, the logistics operation device intelligently acquires a resource distribution request for the second intelligent logistics distribution environment and sends the resource distribution request for the second intelligent logistics distribution environment to the artificial intelligence server, after receiving the resource distribution request for the second intelligent logistics distribution environment, the artificial intelligence server acquires a large number of marked first distribution resource distribution information examples from the information sample set, but the first distribution resource distribution information examples adopt first intelligent logistics distribution environments different from the second intelligent logistics distribution environment, and the artificial intelligence neural network is started to identify and process the plurality of first distribution resource distribution information examples so as to acquire a plurality of second distribution resource distribution information examples which are matched with the plurality of first distribution resource distribution information examples in a one-to-one manner. Wherein the type tag of the second dispatch resource distribution information example is based on the type tag of the first dispatch resource distribution information example. In this way, the type marking of the distributed information example of the distribution resource can be quickly completed without manual marking. And further, the distribution resource scheduling deviation caused by manual marking can be improved to a certain extent.
Step 220, training a first resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second intelligent logistics distribution environment and corresponding type differentiation keywords respectively.
In a specific implementation process, the type-differentiation keywords may be understood as type-differentiation labels, and the first resource distribution differentiation analysis network may be understood as a model for performing classification processing on resource distribution.
Exemplarily, after the artificial intelligence server receives a resource distribution request of a second smart logistics distribution environment, a part of marked third distribution resource distribution information examples are acquired from the information sample set, and the first resource distribution differentiation analysis network is trained through a plurality of third distribution resource distribution information examples and corresponding type differentiation keywords, so that the trained first resource distribution differentiation analysis network can perform resource distribution analysis based on the second smart logistics distribution environment, and logistics resources are scheduled based on a resource distribution analysis result to meet different logistics distribution requirements.
In a possible embodiment, the training of the first resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to the multiple third distribution resource distribution information examples of the second intelligent logistics distribution environment and the corresponding type differentiation keywords in step 220 may specifically include the following steps 2201-2204.
Step 2201, performing an ith round of training on the first resource distribution differentiation analysis network according to the multiple third distribution resource distribution information examples of the second intelligent logistics distribution environment and the corresponding type differentiation keywords respectively.
And 2202, performing ith round information selection processing according to quantitative credible data on the plurality of second distribution resource distribution information examples through the first resource distribution differentiation analysis network trained in the ith round.
Step 2203, performing the (i + 1) th round of training on the first resource distribution differentiation analysis network according to the previous i-round information selection result, the multiple third distribution resource distribution information examples and the corresponding type differentiation keywords respectively.
Step 2204, using the first resource distribution differentiation analysis network of the ith round of training as the first resource distribution differentiation analysis network after completing the training; wherein, i is an integer with constraint condition meeting the condition that i is more than or equal to 1 and less than or equal to j-1, and the value is self-added by one from 1, and j is an integer more than 2 and is used for representing the total accumulated round number of the circular training.
When the content described in steps 2201 to 2204 is specifically implemented, based on a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and the type differentiation keywords respectively corresponding thereto, the first resource distribution differentiation analysis network is subjected to loop iterative training, so that third distribution resource distribution information examples with relatively higher multivalent values are screened out through the first resource distribution differentiation analysis network which is gradually improved, and subsequent continuous improvement training is performed.
In some possible embodiments, the training of the first resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to the multiple third distribution resource distribution information examples of the second intelligent logistics distribution environment and the corresponding type differentiation keywords in step 220 may be further specifically described by the contents recorded in steps 2205-2207.
Step 2205, estimating a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment through the first resource distribution differentiation analysis network, so as to obtain quantitative credible data (confidence) of estimation types (prediction categories) respectively corresponding to the plurality of third distribution resource distribution information examples.
Step 2206, a cost function (loss function) of the first resource distribution differentiation analysis network is formed according to the quantitative credible data of the estimation type and the type differentiation keywords of the third distribution resource distribution information example.
Step 2207, updating parameters (parameters) of the first resource distribution differentiation analysis network until the cost function meets a set condition, and using the updated parameters of the first resource distribution differentiation analysis network when the cost function meets the set condition as the parameters of the trained first resource distribution differentiation analysis network.
When steps 2205 to 2207 are executed, after determining a result of the cost function of the first resource distribution differentiation analysis network based on the quantitative credible data of the estimated type and the type differentiation keyword of the third distributed resource distribution information example, it may be determined whether the result of the cost function of the first resource distribution differentiation analysis network is greater than a set result, when the result of the cost function of the first resource distribution differentiation analysis network is greater than the set result, the first resource distribution differentiation analysis network error warning information is determined based on the cost function of the first resource distribution differentiation analysis network, the error warning information is subjected to feedback processing in the first resource distribution differentiation analysis network, and the network variables of each unit are optimized in the feedback processing process.
In some possible embodiments, the estimating, by the first resource distribution differentiation analysis network, the multiple third distribution resource distribution information examples of the second intelligent logistics distribution environment in step 2205 to obtain quantitative credible data of estimation types corresponding to the multiple third distribution resource distribution information examples, specifically includes the contents described in steps 22051 to 22053: for any one of the plurality of third delivery resource distribution information examples, implementing the following steps: implementing the following steps through the first resource distribution differentiation analysis network: step 22051, performing feature extraction processing (which may be understood as encoding processing) on the third distributed resource distribution information example to obtain a visual key description (encoding vector) of the third distributed resource distribution information example; step 22052, merging the visual key descriptions of the third distributed resource distribution information example to obtain a merged key description; at step 22053, performing non-limiting projection processing (such as mapping processing by an activation function (softmax)) on the merged key description to obtain quantized confidence data (confidence) of the estimation type corresponding to the third distribution resource distribution information example.
When steps 22051-22053 are performed, the first resource distribution differentiation analysis network pair may include an input layer, a hidden layer and an output layer, and the resource distribution differentiation analysis network may be trained rapidly through a part of the third distribution resource distribution information examples, so that the resource distribution differentiation analysis network may perform the resource distribution differentiation analysis task of the second smart logistics distribution environment rapidly. Further, feature extraction processing can be performed on the third distribution resource information example through an input layer to obtain a visual key description of the third distribution resource information example, then merging processing is performed on the visual key description of the third distribution resource information example through a hidden layer to obtain a merged key description, and finally non-limiting projection processing is performed on the merged key description through an output layer to obtain quantitative credible data of an estimation type corresponding to the third distribution resource information example.
In some possible embodiments, the first resource distribution differentiation analysis network comprises a plurality of Sigmoid function units associated with each other. Based on this, the non-limiting projection processing is performed on the merged key description in step 22053, so as to obtain quantitative credible data of an estimation type corresponding to the third distribution resource distribution information example, which may specifically include the following: performing projection operation of a first Sigmoid function unit of the plurality of correlated Sigmoid function units on the merged key description through the first Sigmoid function unit; transmitting the projection result of the first Sigmoid function unit to a downstream Sigmoid function unit associated with each other (the downstream correlation can be understood as subsequent cascade connection), so as to continue projection operation and projection result output in the downstream Sigmoid function unit associated with each other until transmitting to a Sigmoid function unit (the last trigger layer) at the end of an association path; and taking the triggering result output by the Sigmoid function unit positioned at the end of the associated path as quantitative credible data of the estimation type corresponding to the third distribution resource distribution information example.
In the specific implementation of the above, a first Sigmoid function unit passing through a plurality of Sigmoid function units associated with each other may be understood as a first trigger layer passing through a plurality of cascaded trigger layers. Merging key descriptions can be understood as fusing vectors.
Further, the activation function may include W Sigmoid function units (trigger layers), each Sigmoid function unit performs a first-time activation function calculation, the first Sigmoid function unit performs projection of the first Sigmoid function unit on the merging key description to obtain a first projection result, the first projection result is output to a second Sigmoid function unit, the second Sigmoid function unit performs projection of the second Sigmoid function unit on the first projection result to obtain a second projection result until the second Sigmoid result is output to the W Sigmoid function unit, and the trigger result output by the W Sigmoid function unit is used as quantized credible data of an estimation type corresponding to the third distribution resource distribution information example. In this way, the activation function is used for operation, so that the quantitative credible data of the estimation type can be obtained through one-time trigger operation, but the quantitative credible data of the estimation type can be obtained through a plurality of trigger operations, and the difficulty of calculation can be further weakened.
Step 230, performing information selection processing according to quantified credible data on the plurality of second distribution resource distribution information examples through the trained first resource distribution differential analysis network.
In a specific implementation process, after the artificial intelligence server obtains the trained first resource distribution differentiation analysis network through the third distribution resource distribution information example, information selection processing (which can be understood as information screening processing) based on quantitative credible data (which can be understood as confidence) can be performed on a plurality of second distribution resource distribution information examples through the trained first resource distribution differentiation analysis network to select a second distribution resource distribution information example with relatively high available value, and the second resource distribution differentiation analysis network is trained through the second distribution resource distribution information example with relatively high value.
In some possible embodiments, the performing, by the trained first resource distribution differential analysis network, information selection processing according to quantitative credible data on the plurality of second distribution resource distribution information examples in step 230 may specifically include the following: performing the following steps for any one of the plurality of second delivery resource distribution information examples: performing estimation processing (prediction processing) on the second distribution resource information example through the trained first resource distribution differentiation analysis network to obtain quantitative credible data (confidence) of multiple estimation types corresponding to the second distribution resource information example; determining a type differentiation keyword of a first distribution resource distribution information example corresponding to the second distribution resource distribution information example as the type differentiation keyword of the second distribution resource distribution information example; and according to the quantitative credible data of a plurality of estimation types corresponding to the second distribution resource distribution information examples and the type differentiation keywords of the second distribution resource distribution information examples, taking the second distribution resource distribution information examples larger than the quantitative credible data threshold value as second distribution resource distribution information examples obtained by information selection processing.
In specific implementation, the trained first resource distribution differentiation analysis network is used for performing feature extraction processing (which can also be understood as encoding processing) on the second distribution resource distribution information example to obtain a visual key description (encoding vector) of the second distribution resource distribution information example, the visual key descriptions of the second distribution resource distribution information example are combined to obtain a combined key description, the combined key description is subjected to non-limiting projection processing to obtain a plurality of estimation types corresponding to the second distribution resource distribution information example, an estimation type matched with the type differentiation keyword of the second distribution resource distribution information example is determined from the plurality of estimation types corresponding to the second distribution resource distribution information example, and when the quantized credible data of the matched estimation type is greater than a quantized data threshold value, the second distribution resource distribution information example is used as the second distribution resource distribution information example obtained through information selection processing.
Step 240, training a second resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to a second distribution resource distribution information example obtained by the information selection processing.
In an actual implementation process, the architecture complexity of the second resource distribution differentiated analysis network is greater than the architecture complexity of the first resource distribution differentiated analysis network. Wherein, the architecture complexity can be understood as the network depth.
Illustratively, after the artificial intelligence server selects a large number of second distribution resource distribution information examples with relatively high value through the trained first resource distribution differentiation analysis network, the distribution resource distribution information examples of different intelligent logistics distribution environments are generated intelligently, the second resource distribution differentiation analysis network is trained through the large number of second distribution resource distribution information examples with relatively high value, so that the trained second resource distribution differentiation analysis network performs accurate resource distribution analysis based on the second intelligent logistics distribution environment, and the accuracy and timeliness of the resource distribution analysis of the second intelligent logistics distribution environment are improved.
In some possible embodiments, the training of the second resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to the second distribution resource distribution information example obtained by the information selection processing in step 240 may be specifically described by the following three embodiments.
The first embodiment: determining that the second distribution resource distribution information obtained by the information selection processing (screening processing) is exemplified by thermodynamic diagrams (distribution situations) of a plurality of types; when the second distribution resource distribution information example obtained by the information selection processing conforms to thermodynamic diagram coordination indexes in a plurality of types and the number of each type is greater than the corresponding type number threshold value, arbitrarily grabbing the distribution resource distribution information example corresponding to the type number threshold value from the distribution resource distribution information examples of each type in the second distribution resource distribution information example obtained by the information selection processing to form a network training set; and training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the network training set.
In the specific implementation of the first embodiment, the thermodynamic diagrams of the multiple types of the second distributed resource distribution information examples obtained by the determination information selection process may be understood as the distribution of the second distributed resource distribution information examples obtained by the determination screening process in multiple categories. The thermodynamic diagram coordination index can be understood as a distribution equilibrium condition.
Further, after the artificial intelligence server acquires a large number of second distribution resource distribution information examples used for training a second resource distribution differentiation analysis network, thermodynamic diagrams (distribution conditions) of the second distribution resource distribution information examples obtained through the information selection processing in multiple types are analyzed to determine whether the thermodynamic diagrams (distribution conditions) are met, and when the distribution of the second distribution resource distribution information examples obtained through the information selection processing in the multiple types meets the thermodynamic diagrams (distribution equilibrium conditions) and the number of each type is larger than a type number threshold value, the distribution resource distribution information examples of the corresponding type number threshold value are captured (extracted) from the distribution resource distribution information examples of each type in the second distribution resource distribution information examples obtained through the information selection processing to generate a network training set. In this way, the accuracy of the resource distribution differentiation analysis can be improved to a certain extent.
The second embodiment: when the thermodynamic diagrams of the second distribution resource distribution information examples obtained by the information selection processing do not accord with the thermodynamic diagram coordination indexes in the multiple types, performing extended optimization according to the description of the neighbor resources aiming at the second distribution resource distribution information examples of the corresponding types so that the thermodynamic diagrams of the second distribution resource distribution information examples obtained by the extended optimization accord with the thermodynamic diagram coordination indexes in the multiple types; forming a network training set according to the second distribution resource distribution information example obtained by the expansion optimization; and training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the network training set.
In a specific implementation of the second embodiment, the neighbor resource description may be understood as a resource description that is close to the second distributed resource distribution information example, or may be understood as a resource description that has an association relationship with the second distributed resource distribution information example, or may be understood as a resource description that has an association and is relatively similar to the second distributed resource distribution information example.
Further, when the number of the second distribution resource distribution information examples obtained by the information selection processing (screening processing) is smaller than the corresponding number threshold value in each type, performing extended optimization according to the neighbor resource description on the second distribution resource distribution information examples of the corresponding type, so that the number of the second distribution resource distribution information examples obtained by the extended optimization in each category is larger than the corresponding category number threshold value, and generating a network training set based on the second distribution resource distribution information examples obtained by the extended optimization.
The third embodiment: and forming a network training set according to the plurality of third distribution resource distribution information examples and a second distribution resource distribution information example obtained by the information selection processing, and training a second resource distribution differential analysis network used for the second intelligent logistics distribution environment according to the network training set.
In some possible embodiments, the network training set is formed according to the multiple third distribution resource distribution information examples and the second distribution resource distribution information examples obtained through the information selection processing in the third embodiment, and may specifically include the following descriptions: and traversing each type of the second distribution resource distribution information example obtained by the information selection processing, and implementing the following steps: when the number of second distribution resource distribution information examples in the type is lower than a type number threshold of the type, adding a third distribution resource distribution information example arbitrarily grabbed from the plurality of third distribution resource distribution information examples to the second distribution resource distribution information example of the type to update the second distribution resource distribution information example obtained by the information selection processing; and forming a network training set according to a second distribution resource distribution information example obtained by the updated information selection processing.
In embodying the above, when the distribution information examples of some section type are relatively small or when the distribution of some section type is inconsistent, it is possible to add by the third distribution resource distribution information example here. Such as: when the number of second delivery resource distribution information examples in the type is lower than the type number threshold of the type, it indicates that the number of delivery resource distribution information examples in the type is relatively small, and a third delivery resource distribution information example of the type arbitrarily extracted from a plurality of third delivery resource distribution information examples may be added (also may be understood as a supplement) to the second delivery resource distribution information example of the type to update the second delivery resource distribution information example obtained by the information selection processing, so that the delivery resource distribution information examples of the type in the second delivery resource distribution information example can be made richer.
In some possible embodiments, before the training of the second resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to the second distributed resource distribution information example obtained by the information selection processing in step 240, the method may further specifically include the following contents recorded in steps 310 and 320.
In step 310, a target number of examples (target sample number) matching the network performance quantization data that can be used for training the second resource distribution differentiation analysis network is determined according to a mapping list (corresponding relationship) of the network performance quantization data of the resource distribution differentiation analysis network and the number of distributed resource distribution information examples that can be digested (calculated) within a set time sequence step.
And 320, selecting distribution information examples corresponding to the target example number from a network training set consisting of second distribution resource distribution information examples obtained according to the information selection processing as examples for training a second resource distribution differentiated analysis network used for the second intelligent logistics distribution environment.
When the step 310 and the step 320 are executed, the network performance quantitative data of the resource distribution differentiation analysis network can be understood as the network operational capability of the resource distribution differentiation analysis network. The setting of the timing step can be understood as a unit time set in advance.
Further, according to a mapping list (corresponding relation) of the network computing capacity of the resource distribution differentiation analysis network and the number of the distribution resource information examples which can be computed in unit time, the number of target examples which are matched with network performance quantitative data which can be used for training the second resource distribution differentiation analysis network is determined, and distribution resource distribution information examples of the corresponding target examples are selected from a network training set which is composed of the second distribution resource distribution information examples obtained based on information selection processing, so as to serve as examples for training the second resource distribution differentiation analysis network used for the second smart logistics distribution environment. Therefore, the problem of overfitting caused by training the second resource distribution differential analysis network through a large number of training samples can be solved.
In summary, a second distribution resource distribution information example of a second intelligent logistics distribution environment different from a first intelligent logistics distribution environment is obtained through an artificial intelligent neural network, and information selection is carried out on the second distribution resource distribution information example through a first resource distribution differentiation analysis network, so that distribution resource distribution information examples of different intelligent logistics distribution environments are obtained intelligently, and distribution resource scheduling deviation caused by lack of the distribution resource distribution information examples is weakened; furthermore, a second resource distribution differentiation analysis network is trained through a distribution resource distribution information example with relatively high value quantity obtained through information selection, so that the second resource distribution differentiation analysis network can carry out accurate real-time resource distribution differentiation analysis on distribution resource distribution information, the accuracy and timeliness of the differentiation analysis on the distribution resource distribution information are further improved, various logistics distribution demands can be flexibly and accurately met through the second resource distribution differentiation analysis network, and the differentiation analysis result of the distribution resource distribution information is matched with the various logistics distribution demands to achieve rapid, flexible and accurate logistics resource scheduling.
On the basis, please refer to fig. 3, the present application further provides a block diagram of a logistics information processing apparatus 300 based on big data, where the apparatus includes the following functional modules:
an information example obtaining module 310, configured to enable an artificial intelligence neural network according to a plurality of first distribution resource distribution information examples of a first intelligent logistics distribution environment to obtain a plurality of second distribution resource distribution information examples that are one-to-one matched with the plurality of first distribution resource distribution information examples; wherein the plurality of second distribution resource distribution information examples employ a second intelligent logistics distribution environment different from the first intelligent logistics distribution environment;
a first network training module 320, configured to train a first resource distribution differentiation analysis network for the second smart logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and corresponding type differentiation keywords, respectively;
an information example selecting module 330, configured to perform information selection processing according to quantized trusted data on the plurality of second distributed resource distribution information examples through the trained first resource distribution differential analysis network;
a second network training module 340, configured to train a second resource distribution differentiation analysis network for the second smart logistics distribution environment according to a second distribution resource distribution information example obtained by the information selection processing; wherein the architecture complexity of the second resource distribution differentiated analysis network is greater than the architecture complexity of the first resource distribution differentiated analysis network.
Further, a readable storage medium is provided, on which a program is stored, which when executed by a processor implements the method described above.
It will be understood that the present application is not limited to the precise arrangements that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (10)

1. A logistics information processing method based on big data is characterized by being applied to an artificial intelligence server and comprising the following steps:
enabling an artificial intelligence neural network according to a plurality of first distribution resource distribution information examples of a first intelligent logistics distribution environment to obtain a plurality of second distribution resource distribution information examples which are matched with the plurality of first distribution resource distribution information examples in a one-to-one mode; wherein the plurality of second distribution resource distribution information examples employ a second intelligent logistics distribution environment different from the first intelligent logistics distribution environment; wherein the artificial intelligence neural network is a feed-forward neural network; the first intelligent logistics distribution environment is in different distribution states or different distribution flows and comprises logistics sorting environment information, logistics warehousing link information or logistics classification link information; the distribution information of the distributed resources is the configuration conditions of different distributed resources, including the manual distribution condition, the vehicle distribution condition or the production line distribution condition;
training a first resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second intelligent logistics distribution environment and corresponding type differentiation keywords respectively; the type differentiation keywords are type differentiation labels, and the third distribution resource distribution information example is marked in an information sample set; the first resource distribution differentiation analysis network is a model for classifying resource distribution;
selecting and processing the plurality of second distribution resource distribution information examples according to information of quantitative credible data through the trained first resource distribution differential analysis network;
training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to a second distribution resource distribution information example obtained by the information selection processing; wherein the architecture complexity of the second resource distribution differentiated analysis network is greater than the architecture complexity of the first resource distribution differentiated analysis network; the trained second resource distribution differentiation analysis network is used for resource distribution analysis based on a second intelligent logistics distribution environment.
2. The method of claim 1, wherein before training a second resource distribution differentiation analysis network for the second intelligent logistics distribution environment according to a second distribution resource distribution information example obtained by the information selection process, the method further comprises:
determining the number of target examples matched with the network performance quantitative data which can be used for training the second resource distribution differentiated analysis network according to a mapping list of the network performance quantitative data of the resource distribution differentiated analysis network and the number of the distribution resource distribution information examples which can be digested in a set time sequence step;
and selecting the distribution resource information examples corresponding to the target example number from a network training set consisting of second distribution resource distribution information examples obtained according to the information selection processing as examples for training a second resource distribution differentiation analysis network for the second intelligent logistics distribution environment.
3. The method of claim 1, wherein training a first resource distribution differentiation analysis network for the second smart logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and respectively corresponding type differentiation keywords comprises:
performing an ith round of training on the first resource distribution differentiation analysis network according to a plurality of third distribution resource distribution information examples of the second intelligent logistics distribution environment and the type differentiation keywords respectively corresponding to the third distribution resource distribution information examples;
selecting and processing the plurality of second distribution resource distribution information examples according to ith round information of quantized credible data through the first resource distribution differentiation analysis network trained in the ith round;
performing (i + 1) th round training on the first resource distribution differentiation analysis network according to the previous i-round information selection result, the third distribution resource distribution information examples and the corresponding type differentiation keywords respectively;
taking the first resource distribution differential analysis network of the ith round of training as the first resource distribution differential analysis network after completing the training; wherein i is an integer with constraint conditions meeting the condition that i is more than or equal to 1 and less than or equal to j-1, and the value is added by one from 1, and j is an integer more than 2 and is used for representing the total cumulative round number of the cyclic training.
4. The method of claim 1, wherein training a first resource distribution differentiation analysis network for the second smart logistics distribution environment according to a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment and respectively corresponding type differentiation keywords comprises:
estimating a plurality of third distribution resource distribution information examples of the second smart logistics distribution environment through the first resource distribution differentiation analysis network to obtain quantitative credible data of estimation types corresponding to the plurality of third distribution resource distribution information examples respectively;
forming a cost function of the first resource distribution differentiation analysis network according to the quantitative credible data of the estimation type and the type differentiation keywords of the third distribution resource distribution information example;
updating parameters of the first resource distribution differential analysis network until the cost function meets a set condition, and taking the updated parameters of the first resource distribution differential analysis network when the cost function meets the set condition as the parameters of the trained first resource distribution differential analysis network;
correspondingly, the estimating, by the first resource distribution differentiation analysis network, the multiple third distribution resource distribution information examples of the second smart logistics distribution environment to obtain quantitative credible data of estimation types corresponding to the multiple third distribution resource distribution information examples, includes:
for any of the plurality of examples of third dispatch resource distribution information, performing the following steps:
implementing the following steps through the first resource distribution differentiation analysis network:
performing feature extraction processing on the third distribution resource distribution information example to obtain a visual key description of the third distribution resource distribution information example;
merging the visual key descriptions of the third distribution resource distribution information example to obtain a merged key description;
and performing non-limiting projection processing on the merged key description to obtain quantitative credible data of an estimation type corresponding to the third distribution resource distribution information example.
5. The method of claim 4, wherein the first resource distribution differential analysis network comprises a plurality of Sigmoid function units associated with each other; the performing non-limiting projection processing on the merged key description to obtain quantitative credible data of an estimation type corresponding to the third distribution resource distribution information example includes:
performing projection operation of a first Sigmoid function unit of the plurality of related Sigmoid function units on the merged key description through the first Sigmoid function unit;
transmitting the projection result of the first Sigmoid function unit to downstream Sigmoid function units which are mutually associated, and continuing to perform projection operation and projection result output in the downstream Sigmoid function units which are mutually associated until the projection result is transmitted to the Sigmoid function unit positioned at the tail of an associated path;
and taking the trigger result output by the Sigmoid function unit positioned at the tail of the associated path as quantitative credible data of an estimation type corresponding to the third distribution resource distribution information example.
6. The method of claim 1, wherein training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the second distribution resource distribution information example obtained by the information selection process comprises:
determining thermodynamic diagrams of a plurality of types of second distribution resource distribution information examples obtained by the information selection processing;
when the second distribution resource distribution information example obtained by the information selection processing conforms to thermodynamic diagram coordination indexes in a plurality of types and the number of each type is greater than the corresponding type number threshold value, arbitrarily grabbing the distribution resource distribution information example corresponding to the type number threshold value from the distribution resource distribution information examples of each type in the second distribution resource distribution information example obtained by the information selection processing to form a network training set;
and training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the network training set.
7. The method of claim 1, wherein training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the second distribution resource distribution information example obtained by the information selection process comprises:
when the thermodynamic diagrams of the second distribution resource distribution information examples obtained by the information selection processing in the multiple types do not accord with the thermodynamic diagram coordination indexes, performing extended optimization according to the neighbor resource description on the second distribution resource distribution information examples in the corresponding types so that the thermodynamic diagrams of the second distribution resource distribution information examples obtained by the extended optimization in the multiple types accord with the thermodynamic diagram coordination indexes;
forming a network training set according to the second distribution resource distribution information example obtained by the expansion optimization;
and training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to the network training set.
8. The method of claim 1, wherein training a second resource distribution differential analysis network for the second intelligent logistics distribution environment according to a second distribution resource distribution information example obtained by the information selection process comprises:
forming a network training set according to the plurality of third distribution resource distribution information examples and a second distribution resource distribution information example obtained by the information selection processing, and training a second resource distribution differentiation analysis network used for the second intelligent logistics distribution environment according to the network training set;
correspondingly, forming a network training set according to the plurality of third distribution resource distribution information examples and the second distribution resource distribution information examples obtained by the information selection processing, including:
traversing each type of the second distribution resource distribution information example obtained by the information selection processing, and implementing the following steps:
when the number of second distribution resource distribution information examples in the type is lower than a type number threshold of the type, adding a third distribution resource distribution information example of the type arbitrarily grabbed from the plurality of third distribution resource distribution information examples into the second distribution resource distribution information examples of the type to update the second distribution resource distribution information examples obtained by the information selection processing;
and forming a network training set according to a second distribution resource distribution information example obtained by the updated information selection processing.
9. The method of claim 1, wherein the performing of the information selection process according to the quantified credible data on the plurality of second distribution resource distribution information examples through the trained first resource distribution differential analysis network comprises:
performing the following steps for any one of the plurality of second dispatch resource distribution information examples:
estimating the second distribution resource distribution information example through the trained first resource distribution differential analysis network to obtain quantitative credible data of multiple estimation types corresponding to the second distribution resource distribution information example;
determining a type differentiation keyword of a first distribution resource distribution information example corresponding to the second distribution resource distribution information example as the type differentiation keyword of the second distribution resource distribution information example;
and according to the quantitative credible data of a plurality of estimation types corresponding to the second distribution resource distribution information examples and the type differentiation keywords of the second distribution resource distribution information examples, taking the second distribution resource distribution information examples larger than the quantitative credible data threshold value as second distribution resource distribution information examples obtained by information selection processing.
10. An artificial intelligence server comprising a processor and a memory; the processor is in communication with the memory, and the processor is configured to read the computer program from the memory and execute the computer program to implement the method of any one of claims 1 to 9.
CN202211250431.4A 2022-10-13 2022-10-13 Logistics information processing method based on big data and artificial intelligence server Active CN115330319B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211250431.4A CN115330319B (en) 2022-10-13 2022-10-13 Logistics information processing method based on big data and artificial intelligence server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211250431.4A CN115330319B (en) 2022-10-13 2022-10-13 Logistics information processing method based on big data and artificial intelligence server

Publications (2)

Publication Number Publication Date
CN115330319A CN115330319A (en) 2022-11-11
CN115330319B true CN115330319B (en) 2023-01-13

Family

ID=83913587

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211250431.4A Active CN115330319B (en) 2022-10-13 2022-10-13 Logistics information processing method based on big data and artificial intelligence server

Country Status (1)

Country Link
CN (1) CN115330319B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108346023A (en) * 2018-02-01 2018-07-31 广东聚晨知识产权代理有限公司 A kind of big data information management system
CN110070325A (en) * 2019-03-25 2019-07-30 北京蒙成科技有限公司 A kind of dispatching method and device of order to be dispensed
CN114708487A (en) * 2022-04-15 2022-07-05 创域智能(常熟)网联科技有限公司 Logistics distribution business information analysis method and server applying big data
CN115063010A (en) * 2022-07-01 2022-09-16 拉扎斯网络科技（上海）有限公司 Method and device for evaluating distribution resources, electronic equipment and storage medium

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108346023A (en) * 2018-02-01 2018-07-31 广东聚晨知识产权代理有限公司 A kind of big data information management system
CN110070325A (en) * 2019-03-25 2019-07-30 北京蒙成科技有限公司 A kind of dispatching method and device of order to be dispensed
CN114708487A (en) * 2022-04-15 2022-07-05 创域智能(常熟)网联科技有限公司 Logistics distribution business information analysis method and server applying big data
CN115063010A (en) * 2022-07-01 2022-09-16 拉扎斯网络科技（上海）有限公司 Method and device for evaluating distribution resources, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN115330319A (en) 2022-11-11

Similar Documents

Publication Publication Date Title
CN114241282B (en) Knowledge distillation-based edge equipment scene recognition method and device
Lv et al. Generative adversarial networks for parallel transportation systems
CN109271539B (en) Image automatic labeling method and device based on deep learning
US20210357767A1 (en) Automated knowledge infusion for robust and transferable machine learning
CN111737432A (en) Automatic dialogue method and system based on joint training model
CN113343092A (en) Content source recommendation updating method based on big data mining and cloud computing service system
CN114510570A (en) Intention classification method and device based on small sample corpus and computer equipment
CN114492601A (en) Resource classification model training method and device, electronic equipment and storage medium
US20240095529A1 (en) Neural Network Optimization Method and Apparatus
CN116956116A (en) Text processing method and device, storage medium and electronic equipment
CN116643877A (en) Computing power resource scheduling method, training method and system of computing power resource scheduling model
CN113032676B (en) Recommendation method and system based on micro-feedback
CN116910567B (en) Online training sample construction method and related device for recommended service
CN116629338B (en) Model data processing method, related device, equipment and storage medium
CN111368060B (en) Self-learning method, device and system for conversation robot, electronic equipment and medium
CN116484105B (en) Service processing method, device, computer equipment, storage medium and program product
CN108829846A (en) A kind of business recommended platform data cluster optimization system and method based on user characteristics
CN115330319B (en) Logistics information processing method based on big data and artificial intelligence server
CN112508178A (en) Neural network structure searching method and device, electronic equipment and storage medium
CN116976283A (en) Language processing method, training method, device, equipment, medium and program product
Joshi et al. Multiagent systems to support networked scientific computing
CN116108195A (en) Dynamic knowledge graph prediction method and device based on time sequence element learning
CN112926952A (en) Cloud computing-combined big data office business processing method and big data server
CN113239272B (en) Intention prediction method and intention prediction device of network management and control system
CN115423016A (en) Training method of multi-task prediction model, multi-task prediction method and device

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