A network service activation method based on large language model and related equipment

Generate the activation instructions of the application programming interface through a large language model, and automatically handle the network service activation process, solving the problem of time-consuming and labor-intensive writing of instructions manually, and improving the efficiency and accuracy of network service activation.

CN118646648BActive Publication Date: 2024-12-03HANGZHOU YAXIN SOFTWARE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411112748.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-03
Estimated Expiration
2044-08-14

AI Technical Summary

Technical Problem

In the existing network service activation process, it is time-consuming and labor-intensive to manually write activation instructions, which is prone to syntax errors or logic problems, resulting in activation failure and reducing activation efficiency.

Method used

Using a network service activation method based on a large language model, by obtaining service specification data, calling pre-trained large language model to generate activation instructions for the application programming interface, automatically calling the interface to perform the editing operation of activation parameters, improving the degree of automation of activation and efficiency.

Benefits of technology

There is no need to manually write activation instructions, which improves the efficiency and accuracy of network service activation, improves the success rate of activation, and reduces manual intervention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118646648B_ABST
    Figure CN118646648B_ABST
Patent Text Reader

Abstract

The present application discloses a network service activation method based on a large language model and related equipment, which relates to the technical field of large language models. The present application calls a pre-trained large language model to process service activation request data, and generates corresponding application programming interface activation instructions for the tag information of the network service to be activated. The activation instruction is used to call the application programming interface corresponding to the tag information, so that the application programming interface performs the editing operation corresponding to the activation parameters, and sends the activation instruction to the corresponding application programming interface, so that the application programming interface realizes the activation of the network service to be activated. Based on this, there is no need for staff to manually write activation instructions and manually call interfaces, which improves the automation and activation efficiency of network service activation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of large language models, and more specifically, to a network service provisioning method based on a large language model and related equipment. Background Art

[0002] In the existing network service activation process, staff are usually required to manually write and configure the network service activation instructions corresponding to the service order based on the customer's service order information, and assign the activation instructions to the corresponding service interface to complete the activation task for a certain network service.

[0003] However, manually writing activation instructions is time-consuming and laborious, and syntax errors or logical problems are prone to occur during the writing process, resulting in activation failures. Instead, the activation instructions need to be rewritten, which greatly reduces the activation efficiency. Therefore, how to improve the efficiency of network service activation has become an urgent problem to be solved. Summary of the invention

[0004] In view of the above problems, this application provides a network service provisioning method and related devices based on a large language model to achieve the purpose of improving the efficiency of network service provisioning. The specific scheme is as follows:

[0005] The first aspect of the present application provides a network service provisioning method based on a large language model, comprising:

[0006] Acquire service activation request data including service specification data, wherein the service specification data at least includes: at least one set of activation information, wherein the activation information includes: tag information of a network service to be activated, and activation parameters corresponding to the tag information;

[0007] Calling a pre-trained large language model to process the service activation request data, respectively generating corresponding application programming interface activation instructions for the tag information, wherein the activation instructions are used to call the application programming interface corresponding to the tag information, so that the application programming interface performs an editing operation corresponding to the activation parameters, wherein the large language model is a model trained based on a service specification data sample and interface data of the application programming interface, wherein the interface data at least includes: an interface definition and an instruction format;

[0008] The activation instruction is sent to the corresponding application programming interface, so that the application programming interface can realize the activation of the network service to be activated.

[0009] In a possible implementation, obtaining service activation request data including service specification data includes:

[0010] Receiving a service order sent by a user terminal, the service order at least comprising: at least one network service to be activated, and activation parameters corresponding to the network service to be activated;

[0011] The service orders are converted into a preset format, and service activation request data including service specification data corresponding to each of the service orders is obtained respectively.

[0012] In a possible implementation, the training process of the large language model includes:

[0013] Acquire an initial large language model, a service specification data sample for training, and interface data of an application programming interface for training, wherein the interface data at least includes: an interface definition, a parameter description, and an instruction format;

[0014] Inputting the service specification data samples for training and the interface data of the application programming interface for training into the initial large language model;

[0015] Sending a training instruction to the initial large language model so that the initial large language model learns the service specification data sample and the interface data to obtain a large language model, wherein the large language model is capable of generating a mapping relationship between the tag information in the service specification data sample and the application programming interface, and generating an activation instruction in the instruction format that conforms to the application programming interface according to the activation parameters in the service specification data sample.

[0016] In a possible implementation, obtaining a service specification data sample for training and interface data of an application programming interface for training includes:

[0017] Acquire the service specification data sample and the interface data of the application programming interface;

[0018] The service specification data samples and the interface data are converted into a data format that conforms to the initial large language model for training, to obtain the service specification data samples for training and the interface data of the application programming interface for training.

[0019] In a possible implementation, the method further includes:

[0020] When the network service and the application programming interface are updated, the service specification data sample corresponding to the updated network service and the interface data of the updated application programming interface are input into the large language model for training.

[0021] In a possible implementation, the method further includes:

[0022] Inputting historical service activation request data including historical service specification data into the large language model for processing to obtain an activation instruction, and sending the activation instruction to the front end;

[0023] When the front end determines that the activation instruction does not conform to the instruction format, the model parameters of the large language model are adjusted, and the training instruction is sent to the large language model until the front end determines that the activation instruction conforms to the instruction format.

[0024] In a possible implementation, the marking information of the network service to be activated at least includes: a service code representing the network service to be activated.

[0025] A second aspect of the present application provides a network service provisioning device based on a large language model, comprising:

[0026] A data acquisition unit, configured to acquire service activation request data including service specification data, wherein the service specification data at least includes: at least one set of activation information, wherein the activation information includes: tag information of a network service to be activated, and activation parameters corresponding to the tag information;

[0027] A data processing unit, used to call a pre-trained large language model to process the service activation request data, and generate corresponding application programming interface activation instructions for the tag information, respectively, wherein the activation instructions are used to call the application programming interface corresponding to the tag information, so that the application programming interface performs an editing operation corresponding to the activation parameters, wherein the large language model is a model trained based on a service specification data sample and interface data of the application programming interface, wherein the interface data at least includes: an interface definition and an instruction format;

[0028] The instruction sending unit is used to send the activation instruction to the corresponding application programming interface, so that the application programming interface can realize the activation of the network service to be activated.

[0029] A third aspect of the present application provides an electronic device, comprising: at least one processor and a memory connected to the processor, wherein:

[0030] The memory is used to store computer programs;

[0031] The processor is used to execute the computer program so that the electronic device can implement any one of the network service provisioning methods based on a large language model.

[0032] A fourth aspect of the present application provides a computer storage medium on which one or more computer programs are stored. When the computer program is executed by a processor, any one of the methods for providing network services based on a large language model is implemented.

[0033] It can be seen from the above technical solution that the network service activation method based on a large language model and related equipment provided in the embodiment of the present application can utilize the pre-trained large language model to generate activation instructions for the network service to be activated according to the service activation request data, call the application programming interface corresponding to the network service to be activated, and enable it to execute the editing operation corresponding to the activation parameters. There is no need for staff to manually write activation instructions and manually call interfaces, which improves the automation level and activation efficiency of network service activation.

[0034] In addition, by learning the interface data of the application programming interface through a large language model, it is possible to automatically write activation instructions that meet the interface definition and instruction format requirements, thereby improving the accuracy of activation instructions and the success rate of activating network services. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0036] Figure 1 A flowchart of a method for implementing a network service provisioning method based on a large language model provided in an embodiment of the present application;

[0037] Figure 2 A system architecture diagram of a method for implementing network service provisioning based on a large language model provided in an embodiment of the present application;

[0038] Figure 3 A signaling diagram for implementing a method for opening a network service based on a large language model provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of the structure of a network service provisioning device based on a large language model provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of the structure of a network service provisioning device based on a large language model provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0042] Network services refer to various services and functions provided on computer networks. Together, they constitute part of the Internet infrastructure and provide users with rich online experiences and functions. Network services may include: communication services, remote services, search engine services, connection services, digital financial services, etc.

[0043] Taking communication services as an example, a mobile phone number or an IP address is often regarded as a user. If a user wants to activate network services such as data packages or voice and video calls, the operator's customer service is often required to manually write and configure the activation instructions of the corresponding network equipment based on the network service information to be activated in the user's service order information. This is inefficient and has a high error rate. With the development of technology, new types of services are constantly increasing, and the manual configuration and activation of network services cannot meet the needs of fast, accurate and flexible activation.

[0044] In order to solve the above problems, the embodiment of the present application proposes a network service provisioning method based on a large language model. The method can be applied to a processor of a provider providing network services, and the processor executes the following steps: Figure 1 The steps shown, Figure 1 A flow chart of a method for implementing a network service provisioning method based on a large language model provided in an embodiment of the present application, the specific process may include:

[0045] Step S110, obtaining service activation request data including service specification data, the service specification data at least including: at least one set of activation information, the activation information including: marking information of the network service to be activated, and activation parameters corresponding to the marking information.

[0046] The service activation request data is information related to the network service requested by the user to be activated. In the embodiment of the present application, the relevant information is service specification data, which includes unique tag information corresponding to each network service to be activated, and activation parameters corresponding to the tag information. It is understandable that each network service has different specifications. For example, the traffic package to be activated includes multiple packages such as package one and package two, and the activation parameters indicate the service specifications and related parameters of the traffic package to be activated, such as rate limit and traffic quota.

[0047] The service activation request data can be obtained through a variety of channels. The processor can communicate directly with the management end or the user end. The user can send the network service to be activated and the activation parameters of the network service to be activated to the processor through the user end. When the management end is handling faults or customer complaints, if necessary, it can also generate a service activation request and send it to the processor for network activation processing.

[0048] The marking information of the network service to be activated includes at least: a service code representing the network service to be activated, such as a service code 1234J representing a network service for ordering traffic package one.

[0049] Step S120 , calling a pre-trained large language model to process the service activation request data, and generating corresponding application programming interface activation instructions for the tag information.

[0050] Step S130, sending the activation instruction to the corresponding application programming interface, so that the application programming interface can realize the activation of the network service to be activated.

[0051] The activation instruction is used to call the application programming interface (API) corresponding to the tag information, so that the application programming interface performs the editing operation corresponding to the activation parameters. The large language model is a model trained based on service specification data samples and interface data of the application programming interface. The interface data at least includes: interface definition and instruction format.

[0052] APIs are predefined functions that aim to provide applications and developers with the ability to access network devices based on certain software or hardware. Through the generated activation instructions, they call the API interface corresponding to the network service to be activated, access the network device through the API interface, query existing service instances, or create new service instances with the same service specifications, edit the parameters of the service instance so that the parameter values ​​of the service instance are the same as those in the service activation request, and satisfy the service activation request.

[0053] In an embodiment of the present application, the activation instruction is generated by a pre-trained large language model (LLM), wherein the large language model is trained based on a service specification data sample and interface data of an API interface.

[0054] It is understandable that the large language model can learn and understand the complexity and diversity of human language through training with a large amount of text data, and then generate, understand and process natural language text. Based on the above characteristics of the large language model, the embodiment of the present application inputs the service specification data sample and API interface data into the large language model for training. The training process may include the following contents:

[0055] An initial large language model, as well as service specification data samples for training and interface data of an application programming interface for training are obtained, wherein the interface data at least includes: interface definition, parameter description and instruction format; the service specification data samples for training and the interface data of the application programming interface for training are input into the initial large language model; a training instruction is sent to the initial large language model so that the initial large language model learns the service specification data samples and the interface data to obtain a large language model, wherein the large language model can generate a mapping relationship between the tag information in the service specification data samples and the application programming interface, and generate an activation instruction in an instruction format that conforms to the application programming interface according to the activation parameters in the service specification data samples.

[0056] The initial large language model can be a model architecture that is being developed for the first time and trained from scratch. Alternatively, you can use an existing stable large language model, such as the GPT series, ChatGPT, BERT, and other models. Training for network service activation can be performed based on this model, which can greatly save the time and computing resources required for model training. In addition, the existing large language model has been verified in multiple tasks and benchmarks. Training and fine-tuning based on this can improve the reliability of the model in performing network service activation tasks.

[0057] Service specification data is the definition of various service data that is configured by the operator to abstract the service, including the identification, name and corresponding parameters of the service specification. For example, the name or identification information of the network service, the name, data type, length and other information of attributes such as the mobile phone or SIM card number.

[0058] The present application embodiment takes the service specification data related to the network service of a mobile phone number as an example for explanation, and the service specification data is as follows:

[0059] {

[0060] "serviceSpecificationName":"5GaccessService",

[0061] "serviceSpecificationId":"1234",

[0062] "attributes":{

[0063] "msisdn":{

[0064] "name": "msisdn",

[0065] "description":"Mobile phone number",

[0066] "dataType":"String",

[0067] "maxLength":"200",

[0068] },

[0069] "imsi":"460001234567890"{

[0070] "name":"imsi",

[0071] "description": "SIM card number",

[0072] "dataType":"String",

[0073] "maxLength":"200",

[0074] }}}

[0075] The service specification data defines the relevant information of the data "msisdn". "msisdn" represents the mobile phone number, and the data type is a non-negative unsigned character, that is, String, and the maximum character length is 200. Similarly, the service specification data also defines the relevant information of the SIM card number. In the service specification data, "imsi" represents the SIM card number, and the data type is a non-negative unsigned character, that is, String, and the maximum character length is 200.

[0076] In the API interface data, the interface definition can define which network devices corresponding to the network services that the interface can access, as well as the access protocol, etc.; the parameter description can include the parameter type and parameter range of the accessible network devices that can open the network services; the instruction format specifies the correct format of the instruction to call the interface. Optionally, in actual applications, if there are more detailed interface data, service specification data, etc., they can all be used to train the initial large language model. In the embodiments of this application, the data source, data form, data type, data volume, etc. of the training data are not uniquely limited.

[0077] In one possible implementation, service specification data samples and interface data of an application programming interface are obtained; the service specification data samples and the interface data are converted into a data format that conforms to the initial large language model for training, thereby obtaining service specification data samples for training and interface data of an application programming interface for training.

[0078] Service specification data samples and interface data may not meet the format requirements of the large language model for training data, resulting in inaccurate model training results, or the formats of service specification data samples and interface data are not unified, increasing the difficulty of model learning. Based on this, the embodiment of the present application converts the format of service specification data samples and interface data into a data format suitable for large language model training, thereby improving the training efficiency and accuracy of the large language model.

[0079] Furthermore, the above service specification data samples and API interface data are input into the initial large language model for learning by the initial large language model. In the embodiment of the present application, the LLM model can be trained using supervised learning or semi-supervised learning methods, and the LLM model can learn how to map the service specification data to the API interface parameters based on the training data, and generate activation instructions that meet the format specifications.

[0080] Specifically, the large language model can learn the service specification data sample and the API interface definition, map the "msisdn" in the service specification data to the API parameter "msisdn", and map the "imsi" to the API parameter "imsi", thereby generating a mapping relationship between the tag information of the network service to be activated and the API interface. The API interface parameters are the parameters edited when the network device of the API interface activates the network service, that is, the activation parameters in the service specification data.

[0081] Therefore, API interface parameters can be obtained from service specification data. If the activation parameters and the tag information of the network service to be activated exist in the service specification data in the form of grouped activation information, the activation parameters can be directly obtained from the service specification data. However, the activation parameters in the service specification data are not grouped according to the corresponding tag information. The large language model also needs to learn the mapping relationship between the network service to be activated or the tag information and various activation parameters. Based on this, the parameters in the service specification data are classified according to the corresponding tag information to obtain the activation parameters corresponding to the tag information of each network service to be activated.

[0082] Therefore, the trained large language model can process the service specification data, first determine the mapping relationship between the tag information and the API interface, and obtain the activation parameters corresponding to the tag information, and further determine the corresponding required attributes and attribute values ​​in the API interface parameters based on the activation parameters as parameters of the activation instruction. The large language model can generate an activation instruction that meets the format requirements based on the instruction format of the learned API interface. The activation instruction can call the AIP interface and enable the network device corresponding to the API interface to perform the activation operation corresponding to the activation parameters or interface parameters, thereby realizing the activation of the network service to be activated. The large language model sends the generated activation instruction to the corresponding API interface, so that the corresponding network device performs the activation operation corresponding to the activation parameters.

[0083] To sum up, the network service activation method based on a large language model provided in the embodiment of the present application can use the pre-trained large language model to process the service activation request data, generate activation instructions for the network service to be activated, call the application programming interface corresponding to the network service to be activated, and enable it to execute editing operations corresponding to the activation parameters. There is no need for staff to manually write activation instructions and manually call interfaces, which improves the degree of automation and activation efficiency of network service activation.

[0084] In addition, by learning the interface data of the application programming interface through a large language model, it is possible to automatically write activation instructions that meet the interface definition and instruction format requirements, thereby improving the accuracy of activation instructions and the success rate of activating network services.

[0085] Next, other possible implementations of the embodiments of the present application are described.

[0086] A possible implementation of obtaining service activation request data containing service specification data includes: receiving a service order sent by a user terminal, the service order including at least: at least one network service to be activated, and activation parameters corresponding to the network service to be activated; converting the service order into a preset format, and obtaining service activation request data containing service specification data corresponding to each service order.

[0087] The service order sent by the user end to activate the network service has different formats due to factors such as different user ends, different sending platforms, different network service types, and different network service names. Therefore, the embodiment of the present application converts the service order into a unified preset format, such as JSON format, to avoid the large language model from being trained again to generate activation instructions for service orders in different formats, thereby reducing the difficulty of training the large language model.

[0088] Another possible implementation method of network service provisioning based on a large language model may also include:

[0089] When the network service and the application programming interface are updated, the service specification data samples corresponding to the updated network service and the interface data of the updated application programming interface are input into the large language model for training.

[0090] It is understandable that in the context of the rapid development of the Internet, network services and API interfaces will inevitably be upgraded or updated. Therefore, in order to improve the processing capabilities of the large language model, the service specification data corresponding to the upgraded or updated network services and the interface data of the API interface can be sent to the large language model for training, so that the large language model can also obtain corresponding information upgrades, so that it can meet the processing of the updated or upgraded service specification data. Based on this, the embodiments of the present application can easily add new service types and network devices to expand its functions and application scope.

[0091] In one possible implementation, the network service activation method based on a large language model also includes: inputting historical service activation request data containing historical service specification data into the large language model for processing to obtain an activation instruction, and sending the activation instruction to the front end; when the front end determines that the activation instruction does not conform to the instruction format, adjusting the model parameters of the large language model and sending training instructions to the large language model until the front end determines that the activation instruction conforms to the instruction format.

[0092] It is understandable that after the training of the large language model is completed, it needs to be verified to avoid deviations in the model training. Optionally, the large language model can use the training data for self-verification. In the embodiment of the present application, the feedback from the front-end developers can be integrated to adjust the accuracy of the large language model.

[0093] The historical service specification data is input as test data into the trained large language model for processing. The large language model sends the generated activation instructions to the front end, and the front-end developer determines whether it conforms to the instruction format of the corresponding API interface. Optionally, it can also determine whether the API interface called by the activation instruction is correct and whether the activation parameters in the activation instruction are complete.

[0094] If the front-end feedback result is that the activation instruction does not conform to the instruction format of the corresponding API interface, the API interface called by the activation instruction is incorrect, etc., the large language model is retrained and the model parameters of the large language model, such as regularization parameters, residual connection parameters, etc., are adjusted. The above verification steps are performed on the large language model at irregular intervals during the training process until the front-end receives feedback that the activation instruction is in compliance with the instruction format and that there is no problem with the activation instruction, and the training of the large language model is stopped.

[0095] Reference Figure 2 , a system architecture diagram of a method for implementing network service provisioning based on a large language model provided in an embodiment of the present application is used to exemplify the practical application of the method for implementing network service provisioning based on a large language model.

[0096] Figure 2In the present invention, a system for implementing a method for network service activation based on a large language model includes: a data receiving module, a network device interface module, a service specification module and an AI activation module, wherein a service order acceptance submodule in the data receiving module can receive a service order sent by a user end, and a module training submodule in the AI ​​activation module can receive model training feedback or training instructions sent by a management end.

[0097] Reference Figure 3 , a signaling diagram of a method for implementing a network service provisioning method based on a large language model is provided in an embodiment of the present application, which illustrates the role of the above modules in implementing the method for implementing a network service provisioning method based on a large language model.

[0098] The data receiving module includes: a service order acceptance submodule and a data format conversion submodule. The service order acceptance submodule is used to interact with the user end and receive the service order sent by the user end. The data format conversion submodule is used to convert the service order into JSON format and send the service activation request in JSON format to the AI ​​activation module.

[0099] The AI ​​activation module includes: a model training submodule, an instruction generation submodule, and an activation interface call submodule. Among them, the model training submodule is used to receive the instructions of "obtaining service specification data" and "obtaining network device API interface data" sent by the model trainer through the management terminal, and instruct the model training submodule to obtain service specification data and network device API interface data from the service specification module and network device interface module respectively.

[0100] The management end sends a "training activation module" instruction to the model training submodule, and the model training submodule performs training based on the obtained service specification data and network device API interface data to obtain a trained AI activation module. It can be understood that the big language model is a specific technology in the field of AI (Artificial Intelligence), and the AI ​​model includes the big language model, so in this example, the AI ​​model can be used instead of the big language model.

[0101] The instruction generation submodule is used to use the trained AI model to generate activation instructions corresponding to the service activation request, and send the activation instructions to the network device interface module through the activation interface calling submodule. The network activation interface submodule of the network device interface module calls the network device API to perform the corresponding activation operation, activate the network service to be activated in the service activation request, and after completing the activation of the network service, the successful "activation completion status" is fed back to the user end through the data receiving module.

[0102] The network device interface module includes: an interface instruction maintenance submodule, a data provision submodule, and a network activation interface submodule. The interface instruction maintenance submodule is used to obtain the activation AIP interface data from the network device, including interface definition, parameter description, etc., and convert it into the format for AI activation module training. When the "obtain network device API interface data" instruction sent by the AI ​​activation module is received, the API interface data is sent to the AI ​​activation module for its model training.

[0103] The service specification module includes: a service specification maintenance submodule and a data provision submodule. The service specification maintenance submodule is used to receive and store definition data or sample data of various service specifications, such as the attribute name, data type, length, etc. of network services. The stored definition data or sample data is converted into a format qualified for AI activation module training and maintained regularly. When the AI ​​activation module needs to perform model training, the stored service specification data is sent to the AI ​​activation module through the data provision submodule for model training.

[0104] In summary, the embodiments of the present application utilize a large language model (LLM model) and AI technology to learn and understand service specification data and AIP interfaces, and intelligently generate activation instructions based on service orders, adapt to different service types and network equipment, realize the automation of network service processes, reduce manual intervention, and provide efficiency and accuracy in service activation.

[0105] In addition, the service order improves the compatibility and flexibility of the embodiments of the present application by converting service order data in various formats into the same format for model processing.

[0106] The following is a description of a network service provisioning device based on a large language model provided in an embodiment of the present application. The network service provisioning device based on a large language model described below and the network service provisioning method based on a large language model described above can refer to each other.

[0107] First, combine Figure 4 , introduces the network service provisioning device based on large language models, such as Figure 4 As shown, the network service provisioning device based on a large language model may include:

[0108] The data acquisition unit 100 is used to acquire service activation request data including service specification data, wherein the service specification data at least includes: at least one set of activation information, wherein the activation information includes: tag information of the network service to be activated, and activation parameters corresponding to the tag information;

[0109] The data processing unit 200 is used to call the pre-trained large language model to process the service activation request data, and generate corresponding application programming interface activation instructions for the tag information, respectively, the activation instructions are used to call the application programming interface corresponding to the tag information, so that the application programming interface performs the editing operation corresponding to the activation parameters, the large language model is a model trained based on the service specification data sample and the interface data of the application programming interface, and the interface data at least includes: interface definition and instruction format;

[0110] The instruction sending unit 300 is used to send the activation instruction to the corresponding application programming interface, so that the application programming interface can realize the activation of the network service to be activated.

[0111] To sum up, the network service activation device based on a large language model provided in the embodiment of the present application can utilize the pre-trained large language model to generate activation instructions for the network service to be activated according to the service activation request data, call the application programming interface corresponding to the network service to be activated, and enable it to execute editing operations corresponding to the activation parameters. There is no need for staff to manually write activation instructions and manually call interfaces, which improves the degree of automation and activation efficiency of network service activation.

[0112] In addition, by learning the interface data of the application programming interface through a large language model, it is possible to automatically write activation instructions that meet the interface definition and instruction format requirements, thereby improving the accuracy of activation instructions and the success rate of activating network services.

[0113] In a possible implementation, the data acquisition unit 100 includes:

[0114] An order receiving subunit, configured to receive a service order sent by a user terminal, wherein the service order includes at least: at least one network service to be activated, and activation parameters corresponding to the network service to be activated;

[0115] The format conversion subunit is used to convert the service order into a preset format, and obtain service activation request data containing service specification data corresponding to each service order.

[0116] In a possible implementation, the training process of the large language model includes:

[0117] An initial model acquisition unit, used to acquire an initial large language model, a service specification data sample for training, and interface data of an application programming interface for training, wherein the interface data at least includes: an interface definition, a parameter description, and an instruction format;

[0118] A training data input unit, used to input the service specification data samples for training and the interface data of the application programming interface for training into the initial large language model;

[0119] A training unit is used to send training instructions to the initial large language model, so that the initial large language model learns the service specification data sample and the interface data to obtain a large language model, and the large language model can generate a mapping relationship between the tag information in the service specification data sample and the application programming interface, and generate an activation instruction in the instruction format that conforms to the application programming interface according to the activation parameters in the service specification data sample.

[0120] In a possible implementation, the initial model acquisition unit includes:

[0121] A data acquisition subunit, used to acquire the service specification data sample and the interface data of the application programming interface;

[0122] The data format conversion subunit is used to convert the service specification data samples and the interface data into a data format that conforms to the initial large language model for training, and obtain the service specification data samples for training and the interface data of the application programming interface for training.

[0123] In a possible implementation, the method further includes:

[0124] An update training unit is used to input the service specification data sample corresponding to the updated network service and the interface data of the updated application programming interface into the large language model for training when the network service and the application programming interface are updated.

[0125] In a possible implementation, the method further includes:

[0126] An instruction sending unit, used for inputting historical service activation request data including historical service specification data into the large language model for processing, obtaining an activation instruction, and sending the activation instruction to the front end;

[0127] The retraining unit is used to adjust the model parameters of the large language model and send the training instruction to the large language model when the front end determines that the activation instruction does not conform to the instruction format, until the front end determines that the activation instruction conforms to the instruction format.

[0128] In a possible implementation, the marking information of the network service to be activated at least includes: a service code representing the network service to be activated.

[0129] The network service provisioning apparatus based on a large language model provided in the embodiment of the present application can be applied to a network service provisioning device based on a large language model.

[0130] Figure 5 The schematic diagram of the network service provisioning device based on a large language model is shown. Figure 5 The structure of the network service provisioning device based on a large language model may include: at least one processor 10 , at least one memory 20 , at least one communication bus 30 and at least one communication interface 40 .

[0131] In the embodiment of the present application, the number of the processor 10 , the memory 20 , the communication bus 30 , and the communication interface 40 is at least one, and the processor 10 , the memory 20 , and the communication interface 40 communicate with each other through the communication bus 30 .

[0132] The processor 10 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0133] The memory 20 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), etc., such as at least one disk memory.

[0134] The memory stores a program, and the processor can call the program stored in the memory, and the program is used to implement each processing flow in the aforementioned network service provisioning method based on a large language model.

[0135] An embodiment of the present application also provides a computer storage medium, which can store a computer program suitable for execution by a processor, and the computer program is used to implement the various processing flows in the aforementioned network service provisioning method based on a large language model.

[0136] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0137] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0138] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0139] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A network service provisioning method based on a large language model, characterized in that: include: Acquire service activation request data including service specification data, wherein the service specification data at least includes: at least one set of activation information, wherein the activation information includes: marking information of a network service to be activated, and activation parameters corresponding to the marking information, wherein the network service to be activated at least includes: a communication service; Calling a pre-trained large language model to process the service activation request data, respectively generating corresponding application programming interface activation instructions for the tag information, wherein the activation instructions are used to call the application programming interface corresponding to the tag information, so that the application programming interface performs an editing operation corresponding to the activation parameters, wherein the large language model is a model trained based on a service specification data sample and interface data of the application programming interface, wherein the interface data at least includes: an interface definition and an instruction format; Sending the activation instruction to the corresponding application programming interface, so that the application programming interface can activate the network service to be activated; The obtaining of service activation request data including service specification data includes: Receiving a service order sent by a user terminal, the service order at least comprising: at least one network service to be activated, and activation parameters corresponding to the network service to be activated; Converting the service order into a preset format, and obtaining service activation request data including service specification data corresponding to each service order; The training process of the large language model includes: Acquire an initial large language model, a service specification data sample for training, and interface data of an application programming interface for training, wherein the interface data at least includes: an interface definition, a parameter description, and an instruction format; Inputting the service specification data samples for training and the interface data of the application programming interface for training into the initial large language model; Sending a training instruction to the initial large language model, so that the initial large language model learns the service specification data sample and the interface data to obtain a large language model, wherein the large language model can generate a mapping relationship between the tag information in the service specification data sample and the application programming interface, and generate an activation instruction in the instruction format that conforms to the application programming interface according to the activation parameters in the service specification data sample; When the network service and the application programming interface are updated, the service specification data sample corresponding to the updated network service and the interface data of the updated application programming interface are input into the large language model for training.

2. The network service provisioning method based on a large language model according to claim 1, characterized in that: Obtain service specification data samples for training and API interface data for training, including: Acquire the service specification data sample and the interface data of the application programming interface; The service specification data samples and the interface data are converted into a data format that conforms to the initial large language model for training, to obtain the service specification data samples for training and the interface data of the application programming interface for training.

3. The network service provisioning method based on a large language model according to claim 1, characterized in that: Also includes: Inputting historical service activation request data including historical service specification data into the large language model for processing to obtain an activation instruction, and sending the activation instruction to the front end; When the front end determines that the activation instruction does not conform to the instruction format, the model parameters of the large language model are adjusted, and the training instruction is sent to the large language model until the front end determines that the activation instruction conforms to the instruction format.

4. The network service provisioning method based on a large language model according to claim 1, characterized in that: The marking information of the network service to be activated at least includes: a service code representing the network service to be activated.

5. A network service provisioning device based on a large language model, characterized in that: include: A data acquisition unit, configured to acquire service activation request data including service specification data, wherein the service specification data at least includes: at least one set of activation information, wherein the activation information includes: marking information of a network service to be activated, and activation parameters corresponding to the marking information, wherein the network service to be activated at least includes: a communication service; A data processing unit, used to call a pre-trained large language model to process the service activation request data, and generate corresponding application programming interface activation instructions for the tag information, respectively, wherein the activation instructions are used to call the application programming interface corresponding to the tag information, so that the application programming interface performs an editing operation corresponding to the activation parameters, wherein the large language model is a model trained based on a service specification data sample and interface data of the application programming interface, wherein the interface data at least includes: an interface definition and an instruction format; An instruction sending unit, used for sending the activation instruction to the corresponding application programming interface, so that the application programming interface can realize the activation of the network service to be activated; Wherein, the data acquisition unit includes: An order receiving subunit, configured to receive a service order sent by a user terminal, wherein the service order includes at least: at least one network service to be activated, and activation parameters corresponding to the network service to be activated; A format conversion subunit, used to convert the service order into a preset format, and obtain service activation request data including service specification data corresponding to each service order; The functional units used for the large language model include: An initial model acquisition unit, used to acquire an initial large language model, a service specification data sample for training, and interface data of an application programming interface for training, wherein the interface data at least includes: an interface definition, a parameter description, and an instruction format; A training data input unit, used to input the service specification data samples for training and the interface data of the application programming interface for training into the initial large language model; a training unit, configured to send a training instruction to the initial large language model, so that the initial large language model learns the service specification data sample and the interface data to obtain a large language model, wherein the large language model can generate a mapping relationship between the tag information in the service specification data sample and the application programming interface, and generate an activation instruction in the instruction format that conforms to the application programming interface according to the activation parameters in the service specification data sample; An update training unit is used to input the service specification data sample corresponding to the updated network service and the interface data of the updated application programming interface into the large language model for training when the network service and the application programming interface are updated.

6. An electronic device, characterized in that: include: at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the network service provisioning method based on a large language model as described in any one of claims 1 to 4.

7. A computer storage medium having one or more computer programs stored thereon, characterized in that: When the computer program is executed by a processor, the network service provisioning method based on a large language model as described in any one of claims 1 to 4 is implemented.