Data analysis model management methods, electronic equipment, and storage media

The proposed data analysis model management method and system address the lack of ML model management in 5G networks by providing integrated interfaces for registration, training, validation, and deployment, enhancing efficiency and stability.

KR103046706B1Active Publication Date: 2026-11-10ZTE CORP
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020247012066
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-12-30
Filing Date
2022-12-19
Publication Date
2026-11-10
Estimated Expiration
2042-12-19

AI Technical Summary

Technical Problem

Current mobile communication technologies lack effective solutions for managing Machine Learning (ML) models in 5G network systems, which are crucial for achieving intelligent analysis and decision-making capabilities.

Method used

A data analysis model management method and system that includes a model management service module with interfaces for model registration, training, validation, deployment, and update, implemented through a model management service interface on the network management side.

Benefits of technology

Enhances the efficiency and stability of ML model management in communication networks, facilitating automated operations and improving data transmission reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112024039817761-PCT00001_ABST
    Figure 112024039817761-PCT00001_ABST
Patent Text Reader

Abstract

An embodiment according to the present application provides a data analysis model management method, an electronic device, and a storage medium, wherein the method comprises the step of acquiring a model management service module at a network management side of a communication network, wherein the model management service module includes at least one model management service interface; and the step of performing a model management operation on a data analysis model according to the at least one model management service interface.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present application claims priority to a Chinese patent application with application number 202111659735.1 filed with the Chinese Intellectual Property Office on December 30, 2021, the entire contents of said application are incorporated into the present application by reference.

[0002] This application relates to the field of wireless communication technology, and, for example, to a data analysis model management method, electronic equipment, and storage media. Background Technology

[0003] Currently, mobile communication technology has entered the era of 5th Generation Mobile Communication Technology (5G), and network slices have become an important component of 5G. A network slice is an instantiated logical network with characteristic network properties composed primarily of network functions, wherein the network functions include network resources that support them, and the logical network satisfies the requirements of characteristic network services, such as ultra-low latency and ultra-high reliability. As communication technology advances, 5G networks face higher requirements for intelligence in addition to characteristics such as high bandwidth, ultra-low latency, and ultra-high reliability. Currently, automation-related technologies in mobile communication can utilize Self-Organized Networks (SON) and Intentionally Driven Networks. 5G network systems can further incorporate Artificial Intelligence (AI) technology and Big Data technology. Based on these technologies, 5G network systems gradually acquire intelligent analysis and decision-making capabilities, and then implement automated operation and management of 5G networks through the automatic execution of the results of intelligent analysis and decision-making. In addition, as the importance of data analysis for system automation management and operation was confirmed in research by the Third Generation Partnership Project (3GPP), the concepts of Network Data Analytics Function (NWDAF) and Management Data Analytics Function (MDAF) were proposed, and research was conducted on system performance optimization, fault resolution, and system resource prediction using these functions.Research on RAN intelligence frameworks has also begun on the Radio Access Network (RAN) side. All intelligence functions within these RAN intelligence frameworks utilize AI and Machine Learning (ML) technologies, and in the case of AI / ML, they are associated with ML models. The approach proposed during the research process involves the network management layer training the ML models required for RAN-side intelligence functions and performing corresponding updates and deployments after training. To meet these requirements, network management needs to provide corresponding ML model management capabilities; however, there are currently no solutions for such model management in the industry. means of solving the problem

[0004] An embodiment according to the present application provides a data analysis model management method, wherein the method comprises,

[0005] A step of acquiring a model management service module on the network management side of a communication network, wherein the model management service module includes at least one model management service interface; and

[0006] It includes the step of performing a model management operation on a data analysis model according to the above model management service interface.

[0007] An embodiment according to the present application further provides electronic equipment, wherein the electronic equipment is,

[0008] One or more processors; and

[0009] Memory configured to store one or more programs; including,

[0010] When the above one or more programs are executed by the above one or more processors, the above one or more processors are configured to implement a method according to any one of the embodiments of the present application.

[0011] An embodiment according to the present application further provides a computer-readable storage medium in which a computer program is stored, said computer program is executed by a processor to implement a method according to any one of the embodiments of the present application. Brief explanation of the drawing

[0012] FIG. 1 is a flowchart of a data analysis model management method provided in an embodiment according to the present application. FIG. 2 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. FIG. 3 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. FIG. 4 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. FIG. 5 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. FIG. 6 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. FIG. 7 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. FIG. 8 is an example diagram of a data analysis model management method provided in an embodiment according to the present application. FIG. 9 is a schematic diagram of the configuration of a data analysis model management device provided in an embodiment according to the present application. FIG. 10 is a schematic diagram of the configuration of an electronic facility provided in an embodiment according to the present application. Specific details for implementing the invention

[0013] It should be understood that the specific embodiments described herein are for the purpose of interpreting the present application and are not intended to limit the present application.

[0014] In the following description, suffixes used to denote elements, such as "module," "part," or "unit," are used merely to facilitate the description of this application and have no special meaning in themselves. Accordingly, "module," "part," or "unit" may be used interchangeably.

[0015] FIG. 1 is a flowchart of a data analysis model management method provided in an embodiment according to the present application, and the embodiment according to the present application may be applied to the case of data analysis model management in a wireless communication scenario, and the method may be performed by a data analysis module management device provided in an embodiment according to the present application, and the device may be implemented in a software and / or hardware manner, and with reference to FIG. 1, the method provided in an embodiment according to the present application specifically includes the following steps.

[0016] Step (110), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0017] Here, the network management side of the communication network may be an equipment side that manages network communication in a wireless communication network, and may be an operation support system layer, a network management layer, a network element management layer, a slice management layer, and a sub-slice management layer, etc., and the model management service module may be a software and hardware module that manages a data analysis model on the communication network management layer side, and the model management service module may include one or more model management service interfaces, and each model management service interface may manage a data analysis model differently.

[0018] In an embodiment according to the present application, the network management side of the communication network may be equipped with a model management service module, and the model management service module may include one or more model management service interfaces, and each model management service interface may have operation rules for different management operations between data analysis models packaged therein.

[0019] Step (120), according to the model management service interface, perform model management operations on the data analysis model.

[0020] Specifically, through a call to the model management service interface, model management operations of the data analysis model can be implemented, and the difficulty of using the data analysis model in a communication network can be reduced.

[0021] An embodiment according to the present application implements a model management operation for a data analysis model by using a model management service interface, by acquiring a model management service module including at least one model management service interface on the network management side of a communication network. By implementing integrated management of a data analysis model in a communication network through the model management service module, the embodiment according to the present application can reduce the difficulty of using the data analysis model, improve model management efficiency, and guarantee the stability of data transmission in the communication network.

[0022] In one embodiment, based on an embodiment according to the application, the model management service interface includes at least a model registration interface, a model training interface, a model validation interface, a model update interface, a model deployment interface, and a model registration information update interface.

[0023] Specifically, the model management service module may be composed of interfaces such as a model registration interface, a model training interface, a model validation interface, a model update interface, a model deployment interface, and a model registration information update interface, and operations such as model registration, model training, model validation, model update, model deployment, and registration information update can be implemented by calling the interfaces, and processing rules for corresponding operations can be pre-set for each interface.

[0024] FIG. 2 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. Referring to FIG. 2, model registration is an example, and the method provided in an embodiment according to the present application specifically includes the following steps.

[0025] Step (210), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0026] Step (220), call the model registration interface to register the data analysis model and save the attribute information of the data analysis model.

[0027] Here, the model registration interface may be a software program interface, and the processing rules of the registered data analysis model may be packaged in the model registration interface, and when the model registration interface is called, it may register information of the data analysis model in a preset location, and the preset location may be a preset storage space on the network management side of the communication network.

[0028] An embodiment according to the present application can implement a data analysis model registration process by calling a model registration interface among model management service modules and storing attribute information of a data analysis model using the model registration interface, and the user can process based on the registered data analysis model.

[0029] In one embodiment, based on an embodiment according to the application, the attribute information includes at least one of a model identifier, a model name, a model function description, a model algorithm type, a model parameter, a model initial hyperparameter, a training data set, and a test data set.

[0030] In an embodiment according to the present application, one or more of the data analysis model identifier, model name, model function description, model algorithm type, model parameters, model initial hyperparameters, training data set, and test data set can be stored to implement the registration of a data analysis model, and an application in a communication network can implement the use of the data analysis model based on the above-described attribute information.

[0031] In an embodiment according to the present application, a data analysis model can be registered by calling a model registration interface, and the registration of the data analysis model can be implemented by storing attribute information of the data analysis model in a pre-set storage space.

[0032] FIG. 3 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. Referring to FIG. 3, model training is an example, and the method provided in an embodiment according to the present application specifically includes the following steps.

[0033] Step (310), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0034] Step (320), according to a training start command, a model training interface is called to perform training on the data analysis model, wherein the training start command includes at least a model identifier or a model name.

[0035] Here, the training start command may be instruction information for controlling a model to perform training, and the training start command may include a model identifier or model name of the data analysis model to be trained, and the model identifier or model name may consist of numbers, alphabets, or special characters.

[0036] Specifically, when a training start command is received, the model identifier or model name included in the training start command is extracted, and training can be performed on the data analysis model corresponding to the model identifier or model name using the model training interface.

[0037] FIG. 4 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. Referring to FIG. 4, the verification of a data analysis model is an example, and the method provided in an embodiment according to the present application specifically includes the following steps.

[0038] Step (410), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0039] Step (420), according to the verification start command, the model verification interface is called to verify the data analysis model, wherein the verification start command includes at least a model identifier or a model name.

[0040] Here, the validation start command may be instruction information for controlling the model to perform training, and the validation start command may include a model identifier or model name of the data analysis model to be validated, and the model identifier or model name may consist of numbers, alphabets, or special characters.

[0041] In an embodiment according to the present application, when a verification start command is received, a model verification interface is called by the model identifier or model name included in the verification start command to obtain and verify a corresponding data analysis model.

[0042] FIG. 5 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. Referring to FIG. 5, the method provided in the embodiment according to the present application specifically includes the following steps, with the example of updating and using an application of a data analysis model.

[0043] Step (510), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0044] Step (520), according to an update start command, the model update interface is called to update the data analysis model corresponding to the application, wherein the update start command includes at least the application and model identifiers to be updated.

[0045] Here, the update start command may be instruction information for controlling a data analysis model to perform an update, and the update start command may include a model identifier of the data analysis model to be updated and an application to be updated, the model identifier may be unique identifier information of the data analysis model to be updated, and the application to be updated may be an application that uses the data analysis model.

[0046] An embodiment according to the application can, upon receiving an update start command, extract the application and model identifiers to be updated from the update start command, call a model update interface using the application and model identifiers to be updated, and update the data analysis model used by the application to be updated by the model update interface.

[0047] FIG. 6 is a flowchart of another data analysis model management method provided in an embodiment according to the present application, and with reference to FIG. 6, the arrangement of the data analysis model is an example, and the method provided in an embodiment according to the present application specifically includes the following steps.

[0048] Step (610), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0049] Step (620), according to a deployment command, a model deployment interface is called to deploy a data analysis model used by the application, wherein the deployment command includes at least an application and model identifier to be deployed.

[0050] Specifically, upon receiving a deployment command, the model deployment interface is called to deploy a data analysis model corresponding to the model identifier to the application to be deployed, thereby enabling the application to be deployed to implement functions according to the data analysis model.

[0051] FIG. 7 is a flowchart of another data analysis model management method provided in an embodiment according to the present application. Referring to FIG. 7, the method provided in an embodiment according to the present application specifically includes the following steps, with the updating of model registration information as an example.

[0052] Step (710), a model management service module is obtained from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0053] Step (720), according to a registration update command, a model registration information update interface is called to update the registration information of the data analysis model, wherein the registration update command includes at least a model identifier and registration information to be updated.

[0054] In an embodiment according to the present application, when a registration update command is received, the model registration information update interface is called according to the model identifier and the registration information to be updated in the registration update command, thereby replacing the registration information to be updated with the original registration information corresponding to the model identifier, and the update of the registration information can be implemented.

[0055] In one embodiment, based on an embodiment according to the application, the method further includes the step of generating notification information according to the result of performing a model management operation and transmitting the notification information to an application corresponding to a data analysis model.

[0056] In an embodiment according to the present application, after the execution of a model management operation is completed, corresponding notification information, such as successful model registration or successful model update, may be generated according to the result of the execution, for example, and the notification information may be transmitted to an application in which a data analysis model is deployed to notify the application to use the data analysis model of the information.

[0057] In one embodiment, based on the embodiment according to the application, the model management service module is located in one of the Operation-Support System (OSS) layer, the network management layer, the network element management layer, the slice management layer, and the sub-slice management layer.

[0058] FIG. 8 is an exemplary diagram of a data analysis model management method provided in an embodiment according to the present application. Referring to FIG. 8, the data analysis model may specifically be an ML model, and the management method provided in an embodiment according to the present application may include the following steps.

[0059] 1) Register (or deploy) the model to the model database (or model store) through the ML model registration (or deployment) interface.

[0060] 2) Perform training on the registered model through the ML model training interface.

[0061] 3) Validate the trained model through the ML model validation interface.

[0062] 4) Deploy the validated model to a specific application through the ML model deployment interface.

[0063] 5) Regularly performing training, validation, and updating of an ML model includes regularly operating an ML model training interface to perform training on the model; calling an ML model validation interface to perform training on the trained model; and updating the trained model to a feature application that has deployed the ML model through an ML model update interface.

[0064] 6) Call the ML model registration information update model as needed to update the model registration information, and the need may include changes to the model algorithm, changes to the training data set and changes to the test data set, etc.

[0065] The management of the above ML model can be implemented by an ML management service module, and the ML management service module includes, but is not limited to, functional interfaces such as an ML model registration interface, an ML model training interface, an ML model validation interface, an ML model update interface, an ML model deployment interface, and an ML model registration information update interface.

[0066] Here, the ML model registration interface provides an ML model registration function. Any consumer capable of providing a RAN intelligence framework, its RAN intelligence function, or any other authorized ML model may register an ML model by calling the ML model registration interface. The registration information includes, but is not limited to, one or more of the following attributes: a model identifier, a model name, a description of the model function, a model algorithm type (or a specific definition of the model algorithm), model parameters, model initial (or default) hyperparameters, a training data set acquisition path, a training data set, a model test data set acquisition path, and a model test data set; after the model registration is completed, corresponding model information is generated and said model information is stored in a model database (or model store); said model information further includes an attribute that stores identifier information of an application using said ML model, and when the model is registered, the value of said attribute is null. The above training data set acquisition path and model test data set acquisition path can also be replaced with specific model training data set and model test data set.

[0067] The ML model training interface provides ML model training capabilities. Any RAN intelligence framework or its RAN intelligence function, a higher-level network management system, a primary network management function, or any other authorized consumer may call the ML model training interface to start training an ML model. When starting model training, the start command includes at least one of a model identifier and a model name. First, based on the model identifier or model name, parameters such as the corresponding model and model initial (or default) hyperparameters, the training data set acquisition path, or the training data set itself are obtained from the model database, and then model training is started based on the obtained parameters. After training is completed, the corresponding information in the model database is updated, and a notification of model training completion may be sent to the consumer that started the training, and the same notification may also be sent to other applications using the model.

[0068] The ML model validation interface provides ML model validation capabilities. Any RAN intelligence framework, its RAN intelligence function, a higher-level network management system, a primary network management function, or any other consumer may call the ML model validation interface to initiate validation of an ML model. When initiating model validation, the start command includes at least one of a model identifier and a model name. First, parameters, such as the path to obtain the corresponding model testing data set, or the model testing data set itself are obtained from the model database according to the model identifier or model name, and then model validation is initiated according to the obtained parameters or data. After validation is completed, the corresponding information in the model database may be updated, and the model validation results may be transmitted to the consumer initiating the model validation, and simultaneously, the same notification may be transmitted to other applications using the model.

[0069] The ML model update interface provides an ML model update function. Any RAN intelligence framework or its RAN intelligence function, a higher-level network management system, a primary network management function, or any other authorized consumer may call the ML model update interface to initiate an update of an ML model. When initiating a model update, the start command includes at least one of a model identifier and a model name, and an application identifier or a list of applications required for the model update. Depending on the model identifier or model name, corresponding model information may be obtained from a model database and the model may be updated. After the update is completed, a notification of model update completion may be sent to the consumer that initiated the model verification, and simultaneously, the same notification may be sent to the application that updated the model.

[0070] The ML model management service module in the embodiment according to the present application may be located in the operation support system layer, network management layer, network element management layer, slice management layer, and sub-slice management layer.

[0071] In another exemplary embodiment, an ML model information model, or an ML model information object class (IOC), or an ML model managed object class (MOC) is defined, and said model may be named MLModel, wherein said ML model information model comprises: an ML model identifier (MLModelID) that uniquely identifies a model; an ML model algorithm definition that defines a model algorithm; ML model parameters that record model parameters, such as initial or default model parameters recorded therein when the model is first registered, and trained model parameters after training the model; model initial (or default) hyperparameters that record the hyperparameters of the model; a training data set acquisition path or the training data set itself; a model test data set acquisition path or the model test data set itself; 7) a model state that records the state of the model (available states include, but are not limited to, not training, training, unvalidated, validated, undeployed, deployed, etc.); 8) a model deployment application list that records the application to which the affiliated model is deployed; It includes, but is not limited to, one of the attributes. The above ML model information model may further include 1) attributes such as a model name, a description of the model function, etc.In the case of model registration, an instance of MLModel is built and completed by the createMOI operation; and the deletion of the model can be completed by deleting the instance of MLModel by the deleteMOI operation; in the case of model training, first, the model training is completed by the model training operation, and then the model parameters obtained through training are written to the MLModel instance and the model state is modified by the modifyMOIAttributes operation; in the case of model deployment, first, the model deployment is completed by the model deployment operation, and then the model state within the MLModel instance is modified by the modifyMOIAttributes operation and the application information to which the affiliated model is deployed is recorded; in the case of updating ML model registration information, the registration information within the MLModel instance is modified by the modifyMOIAttributes operation, and said registration information includes algorithm definitions, model initial (or default) hyperparameters, training data set acquisition path, model test data set acquisition path, etc.

[0072] FIG. 9 is a schematic diagram of the configuration of a data analysis model management device provided in an embodiment according to the present application, and can perform a data analysis model management method provided in any one embodiment according to the present application, and has a functional module corresponding to the method of execution and beneficial effects, and the device may be implemented by software and / or hardware, and specifically includes a management service module (801) and an operation execution module (802).

[0073] The management service module (801) is configured to obtain a model management service module from the network management side of the communication network, wherein the model management service module includes at least one model management service interface.

[0074] The operation execution module (802) is configured to perform model management operations on the data analysis model according to the model management service interface.

[0075] An embodiment according to the present application obtains a model management service module comprising at least one model management service interface on the network management side of a communication network by means of a management service module, and an operation execution module implements a model management operation for a data analysis model using the model management service interface, and an embodiment according to the present application can reduce the difficulty of using the data analysis model, improve model management efficiency, and guarantee the stability of data transmission in the communication network by implementing integrated management of the data analysis model in the communication network through the model management service module.

[0076] In one embodiment, based on an embodiment according to the application, a model management service interface within the device includes at least a model registration interface, a model training interface, a model verification interface, a model update interface, a model deployment interface, and a model registration information update interface.

[0077] In one embodiment, based on an embodiment according to the application, an operation execution module (802) within the device includes a registration unit configured to call the model registration interface to register the data analysis model and to store attribute information of the data analysis model.

[0078] In one embodiment, based on an embodiment according to the application, the attribute information includes at least one of a model identifier, a model name, a model function description, a model algorithm type, a model parameter, a model initial hyperparameter, a training data set, and a test data set.

[0079] In one embodiment, based on an embodiment according to the application, an operation execution module (802) in the device includes a training unit configured to call the model training interface and perform training on the data analysis model in accordance with a training start command, wherein the training start command includes at least a model identifier or a model name.

[0080] In one embodiment, based on an embodiment according to the application, an operation execution module (802) in the device includes a verification unit configured to call the model verification interface to verify the data analysis model according to a verification start command, wherein the verification start command includes at least a model identifier or a model name.

[0081] In one embodiment, based on an embodiment according to the application, an operation execution module (802) in the device includes an update unit configured to call the model update interface and update the data analysis model corresponding to the application in accordance with an update start command, wherein the update start command includes at least an application and model identifier to be updated.

[0082] In one embodiment, based on an embodiment according to the application, an operation execution module (802) in the device includes a deployment unit configured to deploy the data analysis model used by an application by calling the model deployment interface according to a deployment command, wherein the deployment command includes at least an application and model identifier to be deployed.

[0083] In one embodiment, based on an embodiment according to the application, an operation execution module (802) in the device includes a registration information unit configured to call the model registration information update interface to update the registration information of the data analysis model according to a registration update command, wherein the registration update command includes at least a model identifier and registration information to be updated.

[0084] In one embodiment, based on an embodiment according to the application, the device further includes an information notification module configured to generate notification information according to the result of performing the model management operation and to transmit the notification information to an application corresponding to the data analysis model.

[0085] In one embodiment, based on the embodiment according to the application, the model management service module in the device is located in at least one of the operation support system layer, network management layer, network element management layer, slice management layer, and sub-slice management layer.

[0086] FIG. 10 is a schematic diagram of the configuration of an electronic facility provided in an embodiment according to the present application, wherein the electronic facility includes a processor (90), a memory (91), an input device (92), and an output device (93); the number of processors (90) among the electronic facilities may be one or more, with one processor (90) given as an example in FIG. 10; the processor (90), memory (91), input device (92), and output device (93) among the electronic facilities may be connected by a bus or other method, with FIG. 10 showing a bus connection as an example.

[0087] The memory (91) is a computer-readable storage medium and can store software programs, computer-executable programs, and modules, for example, can store modules (management service module (801) and operation execution module (802)) corresponding to the data transmission device in the embodiment according to the present application. The processor (90) can execute the software programs, instructions, and modules stored in the memory (91) to perform various functional applications and data processing of electronic equipment, that is, can implement the above-described data analysis model management method.

[0088] The memory (91) may primarily include a program storage area and a data storage area, wherein the program storage area may store application programs required for at least one function of the operating system; and the data storage area may store data generated by the use of the electronic equipment. Additionally, the memory (91) may include not only high-speed random access memory but also at least one non-volatile memory such as a magnetic storage device, flash memory or other non-volatile solid-state memory. In some examples, the memory (91) may additionally include memory placed remotely from the processor (90), and such remote memory may be connected to the electronic equipment via a network. Examples of the network include, but are not limited to, the Internet, a corporate intranet, a local area network, a mobile communication network, and combinations thereof.

[0089] The input device (92) can receive input numeric or character information and can generate key signal inputs related to user settings and function control of electronic equipment. The output device (93) may include a display device such as a display screen.

[0090] An embodiment according to the present application further provides a storage medium comprising computer-executable instructions, and when the computer-executable instructions are executed by a computer processor, a data analysis model management method is performed, and the method is

[0091] A step of acquiring a model management service module on the network management side of a communication network, wherein the model management service module includes at least one model management service interface; and

[0092] It includes the step of performing a model management operation on a data analysis model according to the above model management service interface.

[0093] Through the description of the embodiments above, those skilled in the art will understand that the present application may be implemented through software and necessary general-purpose hardware, and of course may be implemented through hardware, but in many cases, the former is a more preferred embodiment. Based on this understanding, the technical solution of the present application, or parts contributing to the prior art, may be implemented in the form of a software product, said computer software product may be stored on a computer-readable storage medium such as, for example, a computer floppy disk, ROM (Read-Only Memory), RAM (Random Access Memory), FLASH, a hard disk, or an optical disk, and may include some instructions to enable a computer device (which may be a personal computer, a server, or a network facility, etc.) to perform the method according to various embodiments of the present application.

[0094] In the embodiments of the above device, each unit and module included is distinguished according to functional logic, but is not limited to the aforementioned distinctions and only needs to be capable of implementing corresponding functions. Furthermore, it should be noted that the specific names of each functional unit are merely intended to facilitate mutual distinction and do not limit the scope of protection of the present application.

[0095] A person skilled in the art should understand that all or part of the steps of the methods disclosed in this text, and functional modules / units of systems and facilities, can be performed using software, firmware, hardware, and suitable combinations thereof.

[0096] In hardware embodiments, the division between function modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a single physical component may have multiple functions, or a single function or step may be performed by the cooperation of multiple physical components. Some or all physical components may be performed by software executed by a processor such as a central processing unit, a digital signal processor, or a microprocessor, or by hardware, or by an integrated circuit such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, and the computer-readable medium may include a computer storage medium (or non-transient medium) and a communication medium (or transient medium). As will be known to a person skilled in the art, the term "computer storage medium" includes volatile and non-volatile, removable and non-removable media implemented in any method or technique containing information (e.g., computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM (Electrically Erasable Programmable Read-Only Memory), flash or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), DVD (Digital Versatile Disc) or other discs, magnetic cassettes, magnetic tapes, magnetic discs or other magnetic storage devices, or any other media that can store desired information and be accessed by a computer. Furthermore, a person skilled in the art will understand that communication media generally include computer-readable instructions, data compositions, program modules or other data in modulated data signals such as carrier waves or other transmission mechanisms, and include any information bearer media.

[0097] Although some embodiments of the present application have been described above with reference to the drawings, the scope of the present application is not limited thereto. Any modifications, equivalent substitutions, and improvements made by a person skilled in the art without departing from the scope and substance of the present application are included within the scope of the present application.

Claims

Claim 1 A data analysis model management method applied to the network management side of a communication network, comprising: a step of acquiring a model management service module—wherein the model management service module includes at least one model management service interface, the network management side of the communication network is an equipment side that manages network communication in a wireless communication network, the model management service module is a software and hardware module that manages a data analysis model at the network management layer side of the communication network, and the model management service module is located in at least one of an operation support system layer, a network management layer, a network element management layer, a slice management layer, and a sub-slice management layer—; and a step of performing a model management operation on a data analysis model according to the at least one model management service interface. Claim 2 A data analysis model management method according to claim 1, wherein the at least one model management service interface comprises at least one of a model registration interface, a model training interface, a model validation interface, a model update interface, a model deployment interface, and a model registration information update interface. Claim 3 A data analysis model management method according to claim 2, wherein the step of performing a model management operation on a data analysis model according to at least one model management service interface comprises, if the at least one model management service interface includes the model registration interface, calling the model registration interface to register the data analysis model and storing attribute information of the data analysis model. Claim 4 A data analysis model management method according to claim 3, wherein the attribute information comprises at least one of a model identifier, a model name, a model function description, a model algorithm type, a model parameter, a model initial hyperparameter, a training data set, and a test data set. Claim 5 In claim 2, the step of performing a model management operation on a data analysis model according to at least one model management service interface includes, if the at least one model management service interface includes the model training interface, the step of calling the model training interface according to a training start command to perform training on the data analysis model, wherein the training start command includes at least a model identifier or a model name. Claim 6 In claim 2, the step of performing a model management operation on a data analysis model according to at least one model management service interface includes, when the at least one model management service interface includes the model verification interface, the step of calling the model verification interface according to a verification start command to verify the data analysis model, wherein the verification start command includes at least a model identifier or a model name. Claim 7 In claim 2, the step of performing a model management operation on a data analysis model according to the at least one model management service interface comprises, when the at least one model management service interface includes the model update interface and performs an update on an application to be updated in which the data analysis model is deployed, the step of calling the model update interface according to an update start command to update the data analysis model corresponding to the application to be updated, wherein the update start command includes at least an application to be updated and a model identifier. Claim 8 In claim 2, the step of performing a model management operation on a data analysis model according to at least one model management service interface comprises, when the at least one model management service interface includes the model deployment interface and the data analysis model is deployed to an application to be deployed, the step of calling the model deployment interface according to a deployment command to deploy the data analysis model used by the application to be deployed, wherein the deployment command includes at least an application to be deployed and a model identifier. Claim 9 In claim 2, the step of performing a model management operation on a data analysis model according to at least one model management service interface includes, if the at least one model management service interface includes the model registration information update interface, the step of calling the model registration information update interface according to a registration update command to update the registration information of the data analysis model, wherein the registration update command includes at least a model identifier and registration information to be updated. Claim 10 A data analysis model management method according to claim 1, further comprising the step of generating notification information based on the result of performing the model management operation and transmitting the notification information to an application in which the data analysis model is deployed. Claim 11 A network management side device of a communication network comprising: one or more processors; and a memory configured to store one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement a method according to any one of claims 1 to 10. Claim 12 A computer-readable storage medium having a computer program stored therein, wherein the computer program is executed by a processor to implement a method according to any one of claims 1 to 10. Claim 13 delete

Citation Information

Patent Citations

  • Control device, control method, and program

    JP2021083058A

  • Distributed learning anomaly detector

    US20200364561A1

  • Data analysis method, device and equipment and storage medium

    CN110569288A

  • Configuring network analytics

    WO2021023388A1