CN114780231A - Service providing method, system and medium based on target requirement of Internet of things - Google Patents
Service providing method, system and medium based on target requirement of Internet of things Download PDFInfo
- Publication number
- CN114780231A CN114780231A CN202210302111.2A CN202210302111A CN114780231A CN 114780231 A CN114780231 A CN 114780231A CN 202210302111 A CN202210302111 A CN 202210302111A CN 114780231 A CN114780231 A CN 114780231A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- target
- learning model
- supplier
- demand
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5022—Mechanisms to release resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a service providing method, a system and a medium based on target requirements of the Internet of things. The method comprises the following steps: receiving a target demand issued by a demand party; determining M suppliers associated with the target demand from N suppliers registered by the system; for each supplier in the M suppliers, training an initial machine learning model configured for the corresponding supplier in the system based on a machine learning algorithm configured for the supplier in the system to obtain a trained machine learning model corresponding to the supplier; verifying the trained machine learning model to obtain a target machine learning model; and taking the supplier corresponding to the target machine learning model as a target supplier, and providing services for the target demand by the target supplier based on the target machine learning model.
Description
Technical Field
The present disclosure relates to the field of artificial intelligence technologies, and in particular, to a method, a system, and a medium for providing a service based on a target requirement of the internet of things.
Background
In recent years, as the artificial intelligence technology becomes more and more abundant and sophisticated, the application scenarios of the artificial intelligence technology also become more and more abundant, and the artificial intelligence technology plays an important role in improving the efficiency of the industry, reducing the cost of the industry, increasing the quality of products and the like.
At present, the mutual supply and demand docking mode for the artificial intelligence technology and the entity application scene mainly comprises the following steps: for example, an application requiring party randomly searches and docks an artificial intelligence technology provider in an offline or online mode, and after successful docking, the two parties sign a cooperative agreement and pay corresponding money to research and develop corresponding products. For example, an artificial intelligence technology party actively searches an application demand party, and after a matched application demand party is found, the two parties sign a cooperative agreement, pay corresponding money and research and develop corresponding products.
However, since the research and development and application of the artificial intelligence technology require training, testing and other processes, in general, the above method still has the following three defects that firstly, after the application demander and the artificial intelligence technology provider reach agreement with some cooperation, the method is changed into the offline traditional research and development, training and testing and other modes, and no coherent and complete online artificial intelligence supply and demand docking, training and testing platform support exists. Secondly, after the application demander and the artificial intelligence technology provider reach the agreement about a certain cooperation, no credible, open and fair artificial intelligence technology training and testing platform develops the training test with the public confidence on the developed artificial intelligence technology. Thirdly, because of the lack of a trusted, open and transparent artificial intelligence technology training and testing platform, testing and evaluation of artificial intelligence technologies and products developed offline will likely result in sub-optimal technologies.
Disclosure of Invention
An object of the disclosed embodiments is to provide a new technical solution for service provision based on target requirements of the internet of things.
According to a first aspect of the present disclosure, a service providing method based on target requirements of the internet of things is provided, which is applied to a service providing system based on target requirements of the internet of things, and the method includes:
receiving a target demand issued by a demand party;
determining M suppliers associated with the target demand from among the N suppliers registered by the system; wherein N, M are integers greater than 1, and N is greater than M;
for each supplier of M suppliers, training an initial machine learning model configured for the corresponding supplier in the system based on a machine learning algorithm configured for the supplier in the system, and obtaining a trained machine learning model corresponding to the supplier;
verifying the trained machine learning model to obtain a target machine learning model;
and taking the supplier corresponding to the target machine learning model as a target supplier, and providing services for the target demand by the target supplier based on the target machine learning model.
Optionally, the receiving the target demand issued by the demand side includes:
auditing the registration information of the demander to obtain an auditing result;
providing a first configuration interface under the condition that the audit result shows that the audit is passed;
and receiving the target requirement issued by the demander through the first configuration interface.
Optionally, the method further comprises:
allocating processing resources to corresponding training tasks when an initial machine learning model configured for the supplier at the system is trained based on a machine learning algorithm configured for the supplier at the system; and (c) a second step of,
and releasing the processing resources allocated to the training task under the condition that the training task is finished.
Optionally, the training, based on the machine learning algorithm configured for the supplier at the system, of the initial machine learning model configured for the corresponding supplier at the system includes:
providing training data matching the target requirements;
generating a training sample according to the training data and the corresponding real label;
and training an initial machine learning model configured for the supplier in the system by using the machine learning algorithm configured for the supplier in the system and the training sample to obtain a trained machine learning model corresponding to the supplier.
Optionally, the verifying the trained machine learning model to obtain a target machine learning model includes:
providing test data matching the target requirements;
predicting the test data according to each trained machine learning model to obtain a prediction label of the test data;
obtaining a judgment index value of each trained machine learning model according to the prediction label and the real label of the test data;
and determining the target machine learning model from the trained machine learning model according to the descending order times of the evaluation index values.
Optionally, after obtaining the trained machine learning model corresponding to the supplier, the method further comprises:
storing the trained machine learning model to a preset position; and (c) a second step of,
providing a viewing interface in response to a request to view the trained machine learning model;
and checking the trained machine learning model at the preset position through the checking interface.
Optionally, before the determining, according to the demand information of the target demand, a plurality of suppliers capable of providing services for the target demand, the method further includes:
providing a second configuration interface;
and acquiring the N suppliers configured through the second configuration interface, and a machine learning algorithm and a corresponding initial machine learning model configured for each supplier.
According to a second aspect of the present disclosure, there is provided a service providing system based on target requirements of the internet of things, comprising:
the receiving module is used for receiving the target demand issued by the demand party;
an association module, configured to determine M suppliers associated with the target demand from among N suppliers registered in the system; wherein N, M are integers greater than 1, and N is greater than M;
a training module, configured to train, for each supplier of M suppliers, an initial machine learning model configured for the supplier at the system based on a machine learning algorithm configured for the supplier at the system, and obtain a trained machine learning model corresponding to the supplier;
the verification module is used for verifying the trained machine learning model to obtain a target machine learning model;
and the matching module is used for taking the supplier corresponding to the target machine learning model as a target supplier and providing service for the target demand by the target supplier based on the target machine learning model.
According to a third aspect of the present disclosure, there is also provided a service providing system based on target requirements of the internet of things, which includes the service providing system based on target requirements of the internet of things according to the second aspect of the present disclosure; or, a memory for storing executable instructions and a processor; the processor is configured to operate under control of the instructions to perform a method as set forth in the first aspect of the disclosure.
One advantageous effect of the present disclosure is that according to the method, system and medium of the embodiments of the present disclosure, a demander and a plurality of suppliers are regarded as events in the internet of things, each independent event is called "item" in the system, through the system, the demander can issue a target demand, and according to demand information of the target demand, M suppliers associated with the target demand are determined from N registered suppliers, that is, an "event" line-of-things characteristic is formed in which one demand corresponds to a plurality of supplies. Meanwhile, aiming at each supplier in the M suppliers, based on a machine learning algorithm configured for the supplier in the system, an initial machine learning model configured for the corresponding supplier in the system is trained, a trained machine learning model corresponding to the supplier is obtained, the trained machine learning model is verified, a target machine learning model is obtained, the supplier corresponding to the target machine learning model serves as the target supplier, and the target supplier provides service for the target demand based on the target machine learning model. Namely, the demand side and the supply side can form credible artificial intelligence full life cycle product generation from demand release to technology docking, model training, model testing and model service through the system. Moreover, the artificial intelligence supply technology can be trained and tested on line in a credible, public and fair way. Secondly, the accuracy, the success degree and the reliability of the butt joint of the demand side and the supply side can be improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a hardware configuration diagram of a service providing system based on target requirements of the internet of things according to an embodiment of the present disclosure;
fig. 2 is a flow chart schematic diagram of a service providing method based on target requirements of the internet of things according to an embodiment of the present disclosure;
FIG. 3 is a schematic diagram of a scenario according to an embodiment of the present disclosure;
FIG. 4 is a functional block diagram of a service providing system based on target requirements of the Internet of things according to an embodiment of the present disclosure;
fig. 5 is a hardware configuration diagram of a service providing system based on target requirements of the internet of things according to another embodiment of the present disclosure.
Detailed Description
Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
< hardware configuration >
Fig. 1 shows a hardware configuration of a service providing system 1000 based on a target requirement of the internet of things that can be used to implement an embodiment of the present disclosure.
The service providing system 1000 based on the target requirement of the internet of things can be deployed in a server or a terminal device. That is, the method of the present embodiment may be implemented by a server, may be implemented by a terminal device, or may be implemented by both the server and the terminal device.
In the application of the method of the embodiment in which the terminal device participates in implementation, the interaction may include human-computer interaction. In the application in which the method of the embodiment is implemented with a server, the interaction may include interaction between the server and the terminal device.
As shown in fig. 1, the service providing system 1000 based on the target demand of the internet of things may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, and the like. The processor 1100 may be a central processing unit CPU, a microprocessor MCU, or the like. The memory 1200 includes, for example, a ROM (read only memory), a RAM (random access memory), a nonvolatile memory such as a hard disk, and the like. The interface device 1300 includes, for example, a USB interface, a headphone interface, and the like. Communication device 1400 is capable of wired or wireless communication, for example. The display device 1500 is, for example, a liquid crystal display panel, a touch panel, or the like. The input device 1600 may include, for example, a touch screen, a keyboard, and the like. A user can input/output voice information through the speaker 1700 and the microphone 1800.
Although a plurality of devices are shown in fig. 1 for the service providing system 1000 based on the target requirement of the internet of things, the disclosure may only relate to some of the devices, for example, the service providing system 1000 based on the target requirement of the internet of things only relates to the memory 1200 and the processor 1100.
In an embodiment of the present disclosure, the memory 1200 of the service providing system 1000 based on the target requirement of the internet of things is used for storing instructions for controlling the processor 1100 to execute the method provided by the embodiment of the present disclosure.
In the above description, the skilled person will be able to design instructions in accordance with the disclosed solution. How the instructions control the operation of the processor is well known in the art and will not be described in detail here.
< method examples >
Fig. 2 is a schematic flowchart of a service providing method based on target requirements of the internet of things according to an embodiment of the present disclosure, where the method is applied to a service providing system based on target requirements of the internet of things, and the system may be deployed in a server and a terminal device, as shown in fig. 2, the method may include the following steps S2100 to S2500:
step S2100, receiving the target demand issued by the demand side.
Target requirements refer to specific requirements published by a claimant. For example, the requesting party is client a, and the target requirement issued by client a may be the automatic driving function of implementing an agv (automated Guided vehicle) car.
In this embodiment, the receiving the target demand issued by the demand side in step S2100 may further include steps S2110 to S2130:
and step S2110, auditing the registration information of the demander, and obtaining an auditing result.
In step S2110, the target demand can be issued only by the demand side registered in the system. Also, there are two states of the issued demand, one of which is a seek docking state in which the demand is not docked with the supplier, and the other of which is a docking completed state in which the demand is already docked with the supplier.
In step S2110, an auditor may audit registration information of a demander of the system, where an audit result includes whether the audit is passed or not passed.
And step S2120, providing a first configuration interface when the audit result shows that the audit is passed.
The first configuration interface may be an input box, a drop down menu, or the like. The first configuration interface is used for a demander to issue target requirements.
In this step S2120, in case that the audit is passed, the demander may be regarded as an "event", and the independent event is an "object" in this system, and at the same time, the demand can issue the target demand conveniently through the provided first configuration interface. And on the contrary, under the condition that the audit is not passed, the demander is prohibited from issuing the target demand through the system.
Step S2130, receiving a target requirement issued by the demander through the first configuration interface.
In this step S2130, when the audit is passed, the requirement may be conveniently issued through the provided first configuration interface. Illustratively, the target demand issued by the demander may be to implement an autopilot function for the AGV carts.
After step S2100 is executed to receive the target demand issued by the demand side, the following steps are entered:
step S2200, determining M suppliers related to the target demand from N suppliers registered by the service providing system based on the target demand of the Internet of things; wherein N, M are integers greater than 1, and N is greater than M.
The supplier is the one capable of providing technical supply. In this embodiment, different suppliers may be registered in the system, and only suppliers registered in the system after manual review can issue supply technologies. The published provisioning techniques may be, among others, machine learning algorithms and initial machine learning models. It will be appreciated that each supplier registered through manual review may also be considered as an "event", each individual "event" is an "item" in the present system, and the items are connected with each other through a timeline and internal logic, and a plurality of connected "items" form a network structure, which is beneficial to closely connecting and corresponding different suppliers to the target demand.
Based on this, the method of the embodiment of the present disclosure further includes: providing a second configuration interface; and acquiring the N suppliers configured through the second configuration interface, and a machine learning algorithm and a corresponding initial machine learning model configured for each supplier.
The second configuration interface may be an input box, a drop down menu, or the like.
In a particular embodiment, a registered supplier through manual review may configure the supplier's provisioning information, the supplier's machine learning algorithm, and a corresponding initial machine learning model through a second configuration interface. Illustratively, as shown in FIG. 3, the suppliers registered via manual review include supplier 1, supplier 2, supplier 3 … … and supplier N. At the same time, the provisioning information for each of the providers, the machine learning algorithm for each of the providers, and the corresponding initial machine learning model may be configured via the second configuration interface. In this way, the N independent "events" of supplier 1, supplier 2, supplier 3 … …, supplier N are "objects", the above target requirements are that the independent "event" is also "object" to realize the automatic driving function of AGV cart, the objects "and" object "are connected by time line and internal logic, the multiple connected" objects "form a net structure, which is beneficial to connect and correspond different suppliers to the target requirements closely.
In a specific embodiment, in the case that the artificially checked demander issues the target demand, the N artificially checked suppliers may determine whether to interface with the target demand according to the target demand, and in the case that the target demand is determined to be interfaced, the supplier may be determined as the supplier associated with the target demand. It will be appreciated that the target demand may be associated with M suppliers simultaneously.
In a specific embodiment, in the case that the demand side after the manual review issues the target demand, the system may automatically determine, from the M suppliers, M suppliers associated with the target demand according to the target demand and the supply information of the N suppliers.
Continuing with the above example, as shown in FIG. 3, the M suppliers determined to be associated with the target demand, i.e., implementing the automated driving function for AGV carts, may be supplier 1, supplier 2, and supplier 3.
In this embodiment, the demander may publish the target demand on the system, generate a "list" of the target demand, and the multiple suppliers may tear off the "list" to form an "event" online internet of things characteristic that one target demand corresponds to multiple suppliers.
After determining M suppliers associated with the target demand from among N suppliers registered in the system in execution of step S2200, the method proceeds to:
step S2300, for each of M suppliers, training an initial machine learning model configured for the supplier at the service providing system based on the target demand based on the internet of things based on a machine learning algorithm configured for the supplier at the service providing system based on the target demand based on the internet of things, to obtain a trained machine learning model corresponding to the supplier.
In this embodiment, the system includes a model training module, which may be a model training module that trains an initial machine learning model configured for a supplier based on a machine learning algorithm configured for the supplier, so as to obtain a trained machine learning model corresponding to the supplier.
In this embodiment, in step S2300, for each supplier of M suppliers, training an initial machine learning model configured for the supplier at the service providing system based on the target requirement of the internet of things based on a machine learning algorithm configured for the supplier at the service providing system based on the target requirement of the internet of things, and obtaining a trained machine learning model corresponding to the supplier may further include steps S2310 to S2330:
step S2310, providing training data matching the target requirement.
In step S2310, the training data provided by all the providers are consistent, and the greater the number of training data, the more accurate the training result is, but after the training data reaches a certain number, the more slow the increase of the accuracy of the training result is until the orientation is stable. Here, the number of training data required for the determination of the accuracy of the training result and the data processing cost can be considered.
Continuing with the above example, in the case that the target requirement is to implement the automatic driving function of the AGV cart, the training data matched with the target requirement may be data related to implementing the automatic driving function of the AGV cart, and the training data may be image data or text data, and the embodiment is not limited herein.
Step S2320, generating a training sample according to the training data and the corresponding real label.
In step S2320, the real label corresponding to the training data may be a label that is labeled manually, or a training sample is generated by combining the real label corresponding to the training data after feature extraction and feature combination are performed on the training data according to an artificial intelligence technique to obtain each target feature.
Step S2330, training the initial machine learning model configured for the supplier at the service providing system based on the target demand of internet of things by using the machine learning algorithm and the training sample configured for the supplier at the service providing system based on the target demand of internet of things, and obtaining a trained machine learning model corresponding to the supplier.
Continuing with the above example, in the case where the suppliers associated with the target demand include supplier 1, supplier 2, and supplier 3, the initial machine learning model provided by supplier 1 may be trained based on the machine learning algorithm corresponding to supplier 1 and the training samples to obtain a trained machine learning model corresponding to supplier 1. Based on the machine learning algorithm corresponding to supplier 2 and the training samples, the initial machine learning model provided by supplier 2 is trained, and a trained machine learning model corresponding to supplier 2 is obtained. And a machine learning algorithm and the training sample corresponding to the supplier 3, training the initial machine learning model provided by the supplier 3, and obtaining a trained machine learning model corresponding to the supplier 3.
In one embodiment, in the case where step S2300 is performed to train an initial machine learning model configured by the system for the corresponding supplier based on a machine learning algorithm configured by the system for the supplier, processing resources may also be allocated for the corresponding training task; and releasing the processing resources allocated to the training task when the training task is finished.
In this embodiment, the system includes a data security sandbox, which is a data flow virtual module based on an open source technology, and has allocatable resources and data, and a function of protecting original resources and data from being damaged and enabling a licensed algorithm and model to run therein. In a specific embodiment, the Processing resources provided by the system include, for example and without limitation, a CPU, a memory usage, a disk Input/Output I/O (Input/Output), a network I/O, a Graphics Processing Unit (GPU), and an FPGA.
In this embodiment, the system further includes a resource allocation module, where the resource allocation module automatically allocates the processing resources to the training task, and releases the allocated processing resources when the training task is completed.
Continuing with the above example, in the case where the initial machine learning model provided by supplier 1 is trained based on the machine learning algorithm and the training sample corresponding to supplier 1, the training task is regarded as training task 1 and processing resources are allocated to training task 1, and in the case where training task 1 ends training, the processing resources allocated to training task 1 are released.
Similarly, in the case where the initial machine learning model provided by provider 2 is trained based on the machine learning algorithm corresponding to provider 2 and the training sample, the training task is regarded as training task 2 and processing resources are allocated to training task 2, and in the case where training task 2 ends training, the processing resources allocated to training task 2 are released.
Similarly, in the case where the initial machine learning model provided by the provider 3 is trained based on the machine learning algorithm corresponding to the provider 3 and the training sample, the training task is regarded as the training task 3 and processing resources are allocated to the training task 3, and in the case where the training task 3 ends training, the processing resources allocated to the training task 3 are released.
After step S2300 is executed for each of M suppliers, training an initial machine learning model configured for the corresponding supplier at the system based on a machine learning algorithm configured for the supplier at the system, and obtaining a trained machine learning model corresponding to the supplier, the method proceeds to:
and step S2400, verifying the trained machine learning model to obtain a target machine learning model.
In this embodiment, the system includes a model verification module, where the model verification module is configured to verify the trained machine learning model to obtain a target machine learning model.
In this embodiment, the verifying the trained machine learning model in step S2400 to obtain the target machine learning model may further include steps S2410 to S2440:
step S2410, providing test data matched with the target requirement.
In step S2410, the test data provided by all suppliers are consistent.
In one example, the test data provided by the system to match the target requirement may be some of the above test data. For example, 10 of 100 pieces of training data are used as test data for verifying the trained machine learning model.
In one example, the test data matching the target requirement provided by the system may be the initial data matching the target requirement divided into the training data and the test data. For example, 150 pieces of data out of 200 pieces of initial data are used as training data for training the initial machine learning model, and 50 pieces of data out of 200 pieces of data are used as test data for verifying the trained machine learning model.
In one example, the test data provided by the system that matches the target requirement may also be re-uploaded data that matches the target requirement. For example, 20 pieces of data are re-uploaded as test data for verification of the trained machine learning model.
Step S2420, predicting the test data according to each trained machine learning model, and obtaining a prediction label of the test data.
And step S2430, obtaining the evaluation index value of each trained machine learning model according to the prediction label and the real label of the test data.
In step S2430, the real label of the test data may be a manually labeled label.
The evaluation index value is an index value for measuring the effect of the trained machine learning model. The evaluation index value may be at least one of Recall rate Recall, Accuracy Precision, Accuracy. It can be understood that the higher the evaluation index value, the better the model is.
Continuing with the above example, after obtaining the trained machine learning model 1 corresponding to the supplier 1, the trained machine learning model 2 corresponding to the supplier 2, and the trained machine learning model 3 corresponding to the supplier 3, the evaluation index value of the trained machine learning model 1, the evaluation index value of the trained machine learning model 2, and the evaluation index value of the trained machine learning model 3 may be calculated.
And S2440, determining the target machine learning model from the trained machine learning models according to the descending order sorting times of the evaluation index values.
In step S2440, the system determines a trained machine learning model with the highest evaluation index value from the trained machine learning models according to the descending order of the evaluation index values, and uses the trained machine learning model as the target machine learning model.
Continuing with the above example, the evaluation index values of the trained machine learning model 1, the evaluation index values of the trained machine learning model 2, and the evaluation index values of the trained machine learning model 3 may be ranked from high to low, and the evaluation index value of the trained machine learning model 1 is the highest, indicating that the effect of the trained machine learning model 1 is the best, so that the trained machine learning model 1 may be used as the target machine learning model.
After the trained machine learning model is verified in step S2500 to obtain the target machine learning model, the method proceeds to:
and step S2600, taking the supplier corresponding to the target machine learning model as a target supplier, and providing a service for the target demand by the target supplier based on the target machine learning model.
In this embodiment, after the trained machine learning model with the best effect is determined as the target machine learning model, the supplier corresponding to the target machine learning model may be used as the target supplier, that is, a matching relationship between the target demand and the target supplier is established, so that the target supplier provides a service for the target demand based on the target machine learning model.
Continuing with the above example, using the trained machine learning model 1 as the target machine learning model, a matching relationship between supplier 1 and the target demand can be established to provide services for the target demand by supplier 1 based on the trained machine learning model 1.
According to the method, the system and the medium of the embodiment of the disclosure, a demand party and a plurality of supply parties are regarded as events in the internet of things, each independent event is called as an 'article' in the system, through the system, the demand party can issue a target demand, and according to the demand information of the target demand, M supply parties associated with the target demand are determined from N registered supply parties, namely, an 'event' online internet of things characteristic that one demand corresponds to a plurality of supplies is formed. Meanwhile, aiming at each supplier in the M suppliers, training an initial machine learning model configured for the corresponding supplier in the system on the basis of a machine learning algorithm configured for the supplier in the system, obtaining a trained machine learning model corresponding to the supplier, verifying the trained machine learning model, obtaining a target machine learning model, taking the supplier corresponding to the target machine learning model as the target supplier, and providing service for the target demand by the target supplier on the basis of the target machine learning model. Namely, the demand side and the supply side can form credible artificial intelligence full life cycle product generation from demand release to technology docking, model training, model testing and model service through the system. Moreover, the artificial intelligence supply technology can be trained and tested on line in a credible, public and fair way. And secondly, the accuracy, the success degree and the reliability of the butt joint of the demand side and the supply side can be improved.
In one embodiment, the method of the present disclosure further includes steps S3100 to S3300 of:
and step S3100, storing the trained machine learning model to a preset position.
In step S3100, the system further includes a model service module, which can store all the trained machine learning models to a preset location, so as to facilitate searching.
Step S3200, in response to a request to view the trained machine learning model, provides a viewing interface.
And S3300, checking the trained machine learning model at the preset position through the checking interface.
According to the embodiment, the trained machine learning model can be uniformly stored, and the external interface is provided for viewing.
< apparatus embodiment >
In this embodiment, a service providing system 4000 based on target requirements of the internet of things is further provided, as shown in fig. 4, the system 4000 includes a receiving module 4100, an association module 4200, a training module 4300, a verification module 4400, and a matching module 4500, and is configured to implement the service providing method based on target requirements of the internet of things provided in this embodiment, each module of the service providing apparatus 4000 based on target requirements of the internet of things may be implemented by software, or may be implemented by hardware, which is not limited herein.
A receiving module 4100, configured to receive a target requirement issued by a demander.
An association module 4200, configured to determine M suppliers associated with the target demand from among the N suppliers registered in the system; n, M are integers greater than 1, and N is greater than M.
A training module 4300, configured to train, for each supplier of the M suppliers, an initial machine learning model configured for the corresponding supplier in the system based on a machine learning algorithm configured for the supplier in the system, and obtain a trained machine learning model corresponding to the supplier.
And the verification module 4400 is used for verifying the trained machine learning model to obtain a target machine learning model.
A matching module 4500, configured to use the supplier corresponding to the target machine learning model as a target supplier, and provide a service for the target demand by the target supplier based on the target machine learning model.
In an embodiment, the receiving module 4100 is specifically configured to perform an audit on the registration information of the demander, and obtain an audit result; providing a first configuration interface under the condition that the audit result shows that the audit is passed; and receiving the target requirement issued by the demander through the first configuration interface.
In one embodiment, the apparatus 4000 further comprises a resource allocation module (not shown in the figure).
A resource allocation module, configured to allocate processing resources to corresponding training tasks when an initial machine learning model configured for the corresponding supplier at the system is trained based on a machine learning algorithm configured for the supplier at the system; and releasing the processing resources allocated to the training task when the training task is finished.
In one embodiment, the training module 4300 is specifically configured to provide training data matching the target requirement; generating a training sample according to the training data and the corresponding real label; and training an initial machine learning model configured by the system for the supplier by using the machine learning algorithm configured by the system for the supplier and the training sample to obtain a trained machine learning model corresponding to the supplier.
In one embodiment, the verification module 4300 is specifically configured to provide test data matching the target requirement; predicting the test data according to each trained machine learning model to obtain a prediction label of the test data; obtaining a judgment index value of each trained machine learning model according to the prediction label and the real label of the test data; and determining the target machine learning model from the trained machine learning model according to the descending order sorting times of the evaluation index values.
In one embodiment, the apparatus 4000 further comprises a model service module (not shown).
The model service module is used for storing the trained machine learning model to a preset position; and, in response to a request to view the trained machine learning model, providing a viewing interface; and checking the trained machine learning model at the preset position through the checking interface.
In one embodiment, the apparatus 4000 further comprises a configuration module (not shown).
A configuration module for providing a second configuration interface; and acquiring the N suppliers configured through the second configuration interface, and a machine learning algorithm and a corresponding initial machine learning model configured for each supplier.
It is understood that the above training module 4100 can implement the functions of the model training module described in the above embodiment, the above verification module 4400 can implement the functions of the model verification module described in the above embodiment, and the above storage module can implement the functions of the model service module described in the above embodiment.
< apparatus embodiment >
Corresponding to the foregoing method embodiments, in this embodiment, a service providing system based on target requirements of the internet of things is further provided, which is used to implement the service providing method based on target requirements of the internet of things according to any embodiment of the present disclosure.
As shown in fig. 5, the service providing system 5000 based on the target requirement of the internet of things may include a processor 5100 and a memory 5200, the memory 5200 being configured to store executable instructions; the processor 5100 is configured to operate the service providing system based on the target demand of the internet of things according to the control of the instruction to perform the service providing method based on the target demand of the internet of things according to any embodiment of the present disclosure.
The present invention may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therewith for causing a processor to implement various aspects of the present invention.
The computer-readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as punch cards or in-groove raised structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be interpreted as a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or an electrical signal transmitted through an electrical wire.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
Computer program instructions for carrying out operations of the present invention may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart 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 invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). 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 and/or flowchart illustration, and combinations of blocks in the block diagrams and/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 is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are equivalent.
While embodiments of the present invention have been described above, the above description is illustrative, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.
Claims (10)
1. A service providing method based on target requirements of the Internet of things is applied to a service providing system based on target requirements of the Internet of things, and the method comprises the following steps:
receiving a target demand issued by a demand party;
determining M suppliers associated with the target demand from among the N suppliers registered by the system; wherein N, M are integers greater than 1, and N is greater than M;
for each supplier of M suppliers, training an initial machine learning model configured for the corresponding supplier in the system based on a machine learning algorithm configured for the supplier in the system, and obtaining a trained machine learning model corresponding to the supplier;
verifying the trained machine learning model to obtain a target machine learning model;
and taking the supplier corresponding to the target machine learning model as a target supplier, and providing services for the target demand by the target supplier based on the target machine learning model.
2. The method of claim 1, wherein the receiving the target demand issued by the demand party comprises:
auditing the registration information of the demander to obtain an auditing result;
providing a first configuration interface under the condition that the auditing result shows that the auditing is passed;
and receiving the target requirement issued by the demander through the first configuration interface.
3. The method of claim 1, wherein the method further comprises:
allocating processing resources to corresponding training tasks when an initial machine learning model configured for the supplier at the system is trained based on a machine learning algorithm configured for the supplier at the system; and the number of the first and second groups,
and releasing the processing resources allocated to the training task under the condition that the training task is finished.
4. The method of claim 1, wherein training the initial machine learning model configured at the system for the corresponding supplier based on the machine learning algorithm configured at the system for the supplier comprises:
providing training data matching the target requirements;
generating a training sample according to the training data and the corresponding real label;
and training an initial machine learning model configured by the system for the supplier by using the machine learning algorithm configured by the system for the supplier and the training sample to obtain a trained machine learning model corresponding to the supplier.
5. The method of claim 4, wherein the validating the trained machine learning model to obtain a target machine learning model comprises:
providing test data matching the target requirements;
predicting the test data according to each trained machine learning model to obtain a prediction label of the test data;
obtaining a judgment index value of each trained machine learning model according to the prediction label and the real label of the test data;
and determining the target machine learning model from the trained machine learning model according to the descending order sorting times of the evaluation index values.
6. The method of claim 1, wherein after obtaining the trained machine learning model corresponding to the supplier, the method further comprises:
storing the trained machine learning model to a preset position; and the number of the first and second groups,
providing a viewing interface in response to a request to view the trained machine learning model;
and checking the trained machine learning model at the preset position through the checking interface.
7. The method of claim 1, wherein prior to said determining a plurality of suppliers that can service the target demand based on the demand information for the target demand, the method further comprises:
providing a second configuration interface;
and acquiring the N suppliers configured through the second configuration interface, and a machine learning algorithm and a corresponding initial machine learning model configured for each supplier.
8. A service providing system based on target demand of the internet of things, comprising:
the receiving module is used for receiving the target demand issued by the demand party;
an association module, configured to determine M suppliers associated with the target demand from among N suppliers registered in the system; n, M are integers more than 1, and N is more than M;
a training module, configured to train, for each supplier of M suppliers, an initial machine learning model configured for the supplier at the system based on a machine learning algorithm configured for the supplier at the system, and obtain a trained machine learning model corresponding to the supplier;
the verification module is used for verifying the trained machine learning model to obtain a target machine learning model;
and the matching module is used for taking the supplier corresponding to the target machine learning model as a target supplier and providing service for the target demand by the target supplier based on the target machine learning model.
9. A service providing system based on target demand of Internet of things, comprising: a processor and a memory for storing instructions for controlling the processor to perform the method of any of claims 1 to 7.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210302111.2A CN114780231B (en) | 2022-03-24 | 2022-03-24 | Service providing method, system and medium based on target requirement of Internet of things |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210302111.2A CN114780231B (en) | 2022-03-24 | 2022-03-24 | Service providing method, system and medium based on target requirement of Internet of things |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114780231A true CN114780231A (en) | 2022-07-22 |
CN114780231B CN114780231B (en) | 2023-03-14 |
Family
ID=82425105
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210302111.2A Active CN114780231B (en) | 2022-03-24 | 2022-03-24 | Service providing method, system and medium based on target requirement of Internet of things |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114780231B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190258783A1 (en) * | 2018-02-21 | 2019-08-22 | International Business Machines Corporation | Stolen machine learning model identification |
CN111611240A (en) * | 2020-04-17 | 2020-09-01 | 第四范式(北京)技术有限公司 | Method, apparatus and device for executing automatic machine learning process |
CN112671865A (en) * | 2020-12-16 | 2021-04-16 | 之江实验室 | SDN-based Internet of things microservice automatic arrangement method |
CN112966438A (en) * | 2021-03-05 | 2021-06-15 | 北京金山云网络技术有限公司 | Machine learning algorithm selection method and distributed computing system |
US20210241033A1 (en) * | 2020-01-31 | 2021-08-05 | Salesforce.Com, Inc. | Test suite for different kinds of biases in data |
-
2022
- 2022-03-24 CN CN202210302111.2A patent/CN114780231B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190258783A1 (en) * | 2018-02-21 | 2019-08-22 | International Business Machines Corporation | Stolen machine learning model identification |
US20210241033A1 (en) * | 2020-01-31 | 2021-08-05 | Salesforce.Com, Inc. | Test suite for different kinds of biases in data |
CN111611240A (en) * | 2020-04-17 | 2020-09-01 | 第四范式(北京)技术有限公司 | Method, apparatus and device for executing automatic machine learning process |
CN112671865A (en) * | 2020-12-16 | 2021-04-16 | 之江实验室 | SDN-based Internet of things microservice automatic arrangement method |
CN112966438A (en) * | 2021-03-05 | 2021-06-15 | 北京金山云网络技术有限公司 | Machine learning algorithm selection method and distributed computing system |
Also Published As
Publication number | Publication date |
---|---|
CN114780231B (en) | 2023-03-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10051082B2 (en) | Cost determination to provide software as a service | |
US10929490B2 (en) | Network search query | |
US11250436B2 (en) | Index of usability for a replacement payment card | |
US10748102B2 (en) | Just in time learning driven by point of sale or other data and metrics | |
US11200587B2 (en) | Facilitating use of select hyper-local data sets for improved modeling | |
US10262266B2 (en) | Identifying and analyzing impact of an event on relationships | |
US20200065912A1 (en) | Expense validator | |
US10003559B2 (en) | Aggregating redundant messages in a group chat | |
US11144607B2 (en) | Network search mapping and execution | |
US10429926B2 (en) | Physical object addition and removal based on affordance and view | |
US10733609B2 (en) | Dynamically generated payment token ratings | |
US11206300B2 (en) | Providing instructions during remote viewing of a user interface | |
US20200374356A1 (en) | Push notification management | |
US20180204232A1 (en) | System and method for buying and selling energy | |
CN111506575A (en) | Method, device and system for training branch point traffic prediction model | |
US20180068330A1 (en) | Deep Learning Based Unsupervised Event Learning for Economic Indicator Predictions | |
CN114780231B (en) | Service providing method, system and medium based on target requirement of Internet of things | |
CN110022327B (en) | Short message authentication test method and device | |
US20200135175A1 (en) | Speech-to-text training data based on interactive response data | |
US11558208B2 (en) | Proximity based audio collaboration | |
US10824974B2 (en) | Automatic subject matter expert profile generator and scorer | |
US11288299B2 (en) | Enhanced action fulfillment using classification valency | |
CN112766912A (en) | Method, device and equipment for determining user travel state information | |
US10656986B2 (en) | Dynamically controlling runtime system logging based on end-user reviews | |
US11354540B2 (en) | Feedback prediction in a social media application |
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 |