System, method and device for realizing enterprise service asset intelligent classification processing based on AIGC, processor and readable storage medium

Abstract

The invention relates to a system for realizing intelligent classification processing of enterprise service assets based on AIGC. The system comprises a service asset data acquisition module for automatically extracting service asset API interface information from a code library, a document and a log; the service asset data preprocessing module is used for cleaning, formatting and standardizing the acquired service asset API data; the service asset registration module officially registers the collected service asset API interface information to a service asset library; the service asset intelligent classification module automatically classifies service asset API interfaces by using a natural language processing technology; and the feedback and optimization module is used for collecting user feedback. By adopting the system, the method and the device for realizing the intelligent classification processing of the enterprise service assets based on the AIGC, the processor and the computer readable storage medium thereof, the manual intervention is reduced, and the error rate is reduced; and the burden of a service asset management team is reduced, and the intelligence and automation of enterprise operation are promoted, so that the wide popularization and continuous innovation of digital application are accelerated.

Description

Technical Field

[0001] The present invention relates to the field of computers, and in particular to the field of intelligent classification of enterprise service assets, and specifically refers to a system, method, device, processor and computer-readable storage medium thereof for implementing intelligent classification processing of enterprise service assets based on AIGC. Background Art

[0002] With the rapid advancement of digitalization, the wave of digitalization is sweeping all walks of life around the world, and more business personnel will participate in the construction of enterprise digital applications. Enterprise service asset management is crucial to ensuring the operational efficiency and financial stability of enterprises. It involves the planning, classification, monitoring and evaluation of enterprise service assets to maximize the value of enterprise service assets. However, traditional enterprise service asset management relies on manual operations, which has problems such as inefficiency and error-proneness. At the same time, with the digital transformation, we urgently need an intelligent service asset intelligent classification management capability that supports self-discovery to improve the efficiency of enterprise service asset management and optimization. The enterprise service asset intelligent classification system and method based on AIGC came into being. This method supports automatic discovery of service assets from multiple sources, and realizes intelligent business classification of service assets through AIGC capabilities, as well as log analysis of service asset call frequency and call trend. Finally, through closed-loop feedback and optimization, service assets are positively intelligently classified and managed. The application of AIGC technology provides an efficient and intelligent new model for enterprise service asset management, reduces the burden on service asset management team personnel, improves the participation of business personnel, and promotes enterprise digital transformation and innovation.

[0003] For enterprise service asset management, the mainstream approach in the industry is to use the enterprise service bus (ESB) or API gateway to register, classify, orchestrate, and monitor service assets. Service asset managers first collect the service interfaces that need to be registered on the enterprise service bus (ESB) and register them manually, manually enter the service asset information into the enterprise service bus to form a service asset list; then manually classify the registered service assets and classify and label the enterprise service assets in the form of tags; when the enterprise creates a new digital application or updates some service assets, the enterprise service asset manager needs to manually change the operation on the enterprise service bus again. Through the above methods, service asset managers manually manage enterprise service assets, which is inefficient, error-prone, and heavily dependent on manual operations. With the digital transformation of enterprises, they will face the rapid construction of digital applications, and urgently need an automated and intelligent solution for intelligent classification and management of enterprise service assets. AIGC-based intelligent classification of enterprise service assets is adopted to realize a new model of intelligent management of enterprise service assets, reduce the burden on the service asset management team, improve the participation of business personnel, and promote digital transformation and innovation of enterprises.

[0004] The following is an analysis of the shortcomings of the above technical implementation methods:

[0005] Disadvantage 1: Inefficiency: Enterprise service asset management relies on manual registration and classification, which is inefficient, time-consuming and error-prone, and cannot meet the rapidly changing needs of digital applications.

[0006] Disadvantage 2: Error-prone and manually dependent: Service asset management relies entirely on manual operations, which not only increases the possibility of errors, but also limits the flexibility and scalability of management, making it difficult to adapt to the fast pace of enterprise digital transformation.

[0007] Insufficiency 3: Lack of automation and intelligence: The existing management methods fail to fully utilize advanced technologies such as AIGC to achieve automation and intelligence, resulting in a heavy burden on the service asset management team and low participation of business personnel, hindering the pace of digital transformation and innovation of enterprises. Summary of the invention

[0008] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a system, method, device, processor and computer-readable storage medium thereof for realizing intelligent classification processing of enterprise service assets based on AIGC, which are easy to operate and have a wide range of applications.

[0009] In order to achieve the above objectives, the system, method, device, processor and computer-readable storage medium for implementing intelligent classification processing of enterprise service assets based on AIGC of the present invention are as follows:

[0010] The main features of the system for realizing intelligent classification and processing of enterprise service assets based on AIGC are as follows:

[0011] Service asset data collection module, used to automatically extract service asset API interface information from code bases, documents, and logs;

[0012] A service asset data preprocessing module, connected to the service asset data collection module, is used to clean, format and standardize the collected service asset API data;

[0013] A service asset registration module, connected to the service asset data preprocessing module, is used to receive the output of the service asset data collection module and formally register the collected service asset API interface information into the service asset library;

[0014] A service asset intelligent classification module, connected to the service asset registration module, for automatically classifying service asset API interfaces using natural language processing technology;

[0015] The feedback and optimization module is connected to the service asset intelligent classification module and is used to collect user feedback to optimize system performance and accuracy.

[0016] Preferably, the service asset data collection module automatically extracts service asset API interface information from code bases, documents and logs, specifically including the following steps:

[0017] (1.1) Define the service asset model and define data standard requirements for each field of the service asset model;

[0018] (1.2) Specify the target data source to be collected;

[0019] (1.3) According to the defined target data source, select the corresponding parser and parse the content in the target data source according to the defined service asset model;

[0020] (1.4) Define the collection frequency and collection mode, and automatically execute the collection task;

[0021] (1.5) Verify the format of the collected data to ensure its integrity and accuracy.

[0022] Preferably, the service asset data preprocessing module cleans, formats and standardizes the collected service asset API data, specifically including the following steps:

[0023] (2.1) Remove duplicate data from the data collected by the service asset data collection module;

[0024] (2.2) removing noise and irrelevant information from the data collected by the service asset data collection module;

[0025] (2.3) For each specific service asset API, obtain the call frequency from the log file or log center and perform activity statistics;

[0026] (2.4) All data collected by the service asset data collection module are unified into a standard data format;

[0027] (2.5) The data collected by the service asset data collection module shall be standardized according to the data standards required by the service asset model.

[0028] Preferably, the service asset registration module assigns a unique identification ID to each registered service asset interface, and the service asset registration module registers the service asset API to the service asset library, queries the service asset API details, deletes the service asset API, queries the status information of the service asset API, queries the classification information of the service asset API, updates the service asset API, and updates the calling frequency of the service asset API based on the unique identification ID.

[0029] Preferably, the service asset intelligent classification module uses natural language processing technology to automatically classify the service asset API interface, specifically including the following steps:

[0030] (4.1) Select some data from all the collected service asset API data as training sample sets, and classify and label these sample data. At the same time, set clear activity classification standards based on the call frequency of the service asset API interface, and determine the activity label of each service asset API interface;

[0031] (4.2) Encoding: Perform one-hot encoding conversion on the marked business classification labels and activity labels, and convert the classification labels into a numerical encoding form that can be processed by the machine learning model;

[0032] (4.3) Use deep learning framework to build classification algorithm model;

[0033] (4.4) Use the encoded training sample data to train the model. During the training process, the model performance is optimized by adjusting the model's hyperparameters. At the same time, the model's generalization ability is evaluated to prevent overfitting.

[0034] (4.5) Apply the trained model to the remaining service asset API interface data, perform model reasoning, and automatically classify new service asset API interfaces.

[0035] Preferably, the feedback and optimization module collects user feedback, specifically including the following steps:

[0036] (5.1) Judge the model inference results and classify them into accurate and inaccurate categories;

[0037] (5.2) If it is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set, and the step (5.3) is continued; if it is judged to be accurate, the step is exited;

[0038] (5.3) Continue training the classification algorithm model.

[0039] The method for realizing intelligent classification and processing of enterprise service assets based on AIGC using the above system is mainly characterized in that the method comprises the following steps:

[0040] (1) Service asset data collection module, which automatically extracts service asset API interface information from code bases, documents, and logs;

[0041] (2) The service asset data preprocessing module cleans, formats, and standardizes the collected service asset API data;

[0042] (3) The service asset registration module receives the output of the service asset data collection module and formally registers the collected service asset API interface information into the service asset library;

[0043] (4) The service asset intelligent classification module uses natural language processing technology to automatically classify service asset API interfaces;

[0044] (5) The feedback and optimization module collects user feedback.

[0045] Preferably, the step (1) specifically comprises the following steps:

[0046] (1.1) Define the service asset model and define data standard requirements for each field of the service asset model;

[0047] (1.2) Specify the target data source to be collected;

[0048] (1.3) According to the defined target data source, select the corresponding parser and parse the content in the target data source according to the defined service asset model;

[0049] (1.4) Define the collection frequency and collection mode, and automatically execute the collection task;

[0050] (1.5) Verify the format of the collected data to ensure its integrity and accuracy.

[0051] Preferably, the step (2) specifically comprises the following steps:

[0052] (2.1) Remove duplicate data from the data collected by the service asset data collection module;

[0053] (2.2) removing noise and irrelevant information from the data collected by the service asset data collection module;

[0054] (2.3) For each specific service asset API, obtain the call frequency from the log file or log center and perform activity statistics;

[0055] (2.4) All data collected by the service asset data collection module are unified into a standard data format;

[0056] (2.5) The data collected by the service asset data collection module shall be standardized according to the data standards required by the service asset model.

[0057] Preferably, the step (4) specifically comprises the following steps:

[0058] (4.1) Select some data from all the collected service asset API data as training sample sets, and classify and label these sample data. At the same time, set clear activity classification standards based on the call frequency of the service asset API interface, and determine the activity label of each service asset API interface;

[0059] (4.2) Encoding: Perform one-hot encoding conversion on the marked business classification labels and activity labels, and convert the classification labels into a numerical encoding form that can be processed by the machine learning model;

[0060] (4.3) Use deep learning framework to build classification algorithm model;

[0061] (4.4) Use the encoded training sample data to train the model. During the training process, the model performance is optimized by adjusting the model's hyperparameters. At the same time, the model's generalization ability is evaluated to prevent overfitting.

[0062] (4.5) Apply the trained model to the remaining service asset API interface data, perform model reasoning, and automatically classify new service asset API interfaces.

[0063] Preferably, the step (5) specifically comprises the following steps:

[0064] (5.1) Judge the model inference results and classify them into accurate and inaccurate categories;

[0065] (5.2) If it is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set, and the step (5.3) is continued; if it is judged to be accurate, the step is exited;

[0066] (5.3) Continue training the classification algorithm model.

[0067] The main feature of the device for realizing intelligent classification processing of enterprise service assets based on AIGC is that the device comprises:

[0068] a processor configured to execute computer executable instructions;

[0069] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for realizing intelligent classification processing of enterprise service assets based on AIGC are implemented.

[0070] The processor for realizing intelligent classification and processing of enterprise service assets based on AIGC has the main feature that the processor is configured to execute computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the above-mentioned method for realizing intelligent classification and processing of enterprise service assets based on AIGC are realized.

[0071] The main feature of the computer-readable storage medium is that a computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the above-mentioned method for realizing intelligent classification processing of enterprise service assets based on AIGC.

[0072] The system, method, device, processor and computer-readable storage medium thereof for realizing intelligent classification and processing of enterprise service assets based on AIGC of the present invention have been adopted, and the efficiency and accuracy of service asset management have been significantly improved through automated data collection, preprocessing, registration and intelligent classification; the system uses natural language processing technology to realize automatic classification and activity analysis of service asset API interfaces, reducing manual intervention and lowering the error rate; at the same time, the real-time feedback mechanism of business personnel enables the system to self-learn and optimize, enhancing business participation and improving system performance. The intelligent classification of enterprise service assets based on AIGC has reduced the burden on the service asset management team, promoted the intelligence and automation of enterprise operations, and thus accelerated the widespread popularization and continuous innovation of digital applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 This is a structural diagram of the system for realizing intelligent classification processing of enterprise service assets based on AIGC of the present invention.

[0074] Figure 2 This is a flow chart of the steps of the service asset data collection module of the system for realizing intelligent classification processing of enterprise service assets based on AIGC of the present invention.

[0075] Figure 3This is a flow chart of the steps of the service asset data preprocessing module of the system for realizing intelligent classification processing of enterprise service assets based on AIGC of the present invention.

[0076] Figure 4 This is a flow chart of the steps of the service asset intelligent classification module of the system for realizing enterprise service asset intelligent classification processing based on AIGC of the present invention.

[0077] Figure 5 This is a flow chart of the steps of the feedback and optimization module of the system for realizing intelligent classification processing of enterprise service assets based on AIGC of the present invention. DETAILED DESCRIPTION

[0078] In order to more clearly describe the technical content of the present invention, further description is given below in conjunction with specific embodiments.

[0079] The system for realizing intelligent classification and processing of enterprise service assets based on AIGC of the present invention includes:

[0080] Service asset data collection module, used to automatically extract service asset API interface information from code bases, documents, and logs;

[0081] A service asset data preprocessing module, connected to the service asset data collection module, is used to clean, format and standardize the collected service asset API data;

[0082] A service asset registration module, connected to the service asset data preprocessing module, is used to receive the output of the service asset data collection module and formally register the collected service asset API interface information into the service asset library;

[0083] A service asset intelligent classification module, connected to the service asset registration module, for automatically classifying service asset API interfaces using natural language processing technology;

[0084] The feedback and optimization module is connected to the service asset intelligent classification module and is used to collect user feedback to optimize system performance and accuracy.

[0085] As a preferred embodiment of the present invention, the service asset data collection module automatically extracts service asset API interface information from code bases, documents and logs, specifically including the following steps:

[0086] (1.1) Define the service asset model and define data standard requirements for each field of the service asset model;

[0087] (1.2) Specify the target data source to be collected;

[0088] (1.3) According to the defined target data source, select the corresponding parser and parse the content in the target data source according to the defined service asset model;

[0089] (1.4) Define the collection frequency and collection mode, and automatically execute the collection task;

[0090] (1.5) Verify the format of the collected data to ensure its integrity and accuracy.

[0091] As a preferred embodiment of the present invention, the service asset data preprocessing module cleans, formats and standardizes the collected service asset API data, specifically including the following steps:

[0092] (2.1) Remove duplicate data from the data collected by the service asset data collection module;

[0093] (2.2) removing noise and irrelevant information from the data collected by the service asset data collection module;

[0094] (2.3) For each specific service asset API, obtain the call frequency from the log file or log center and perform activity statistics;

[0095] (2.4) All data collected by the service asset data collection module are unified into a standard data format;

[0096] (2.5) The data collected by the service asset data collection module shall be standardized according to the data standards required by the service asset model.

[0097] As a preferred embodiment of the present invention, the service asset registration module assigns a unique identification ID to each registered service asset interface. The service asset registration module registers the service asset API to the service asset library, queries the service asset API details, deletes the service asset API, queries the status information of the service asset API, queries the classification information of the service asset API, updates the service asset API, and updates the calling frequency of the service asset API based on the unique identification ID.

[0098] As a preferred embodiment of the present invention, the service asset intelligent classification module uses natural language processing technology to automatically classify service asset API interfaces, specifically including the following steps:

[0099] (4.1) Select some data from all the collected service asset API data as training sample sets, and classify and label these sample data. At the same time, set clear activity classification standards based on the call frequency of the service asset API interface, and determine the activity label of each service asset API interface;

[0100] (4.2) Encoding: Perform one-hot encoding conversion on the marked business classification labels and activity labels, and convert the classification labels into a numerical encoding form that can be processed by the machine learning model;

[0101] (4.3) Use deep learning framework to build classification algorithm model;

[0102] (4.4) Use the encoded training sample data to train the model. During the training process, the model performance is optimized by adjusting the model's hyperparameters. At the same time, the model's generalization ability is evaluated to prevent overfitting.

[0103] (4.5) Apply the trained model to the remaining service asset API interface data, perform model reasoning, and automatically classify new service asset API interfaces.

[0104] As a preferred embodiment of the present invention, the feedback and optimization module collects user feedback, specifically comprising the following steps:

[0105] (5.1) Judge the model inference results and classify them into accurate and inaccurate categories;

[0106] (5.2) If it is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set, and the step (5.3) is continued; if it is judged to be accurate, the step is exited;

[0107] (5.3) Continue training the classification algorithm model.

[0108] The method of the present invention for realizing intelligent classification and processing of enterprise service assets based on AIGC using the above system is mainly characterized in that the method comprises the following steps:

[0109] (1) Service asset data collection module, which automatically extracts service asset API interface information from code bases, documents, and logs;

[0110] (2) The service asset data preprocessing module cleans, formats, and standardizes the collected service asset API data;

[0111] (3) The service asset registration module receives the output of the service asset data collection module and formally registers the collected service asset API interface information into the service asset library;

[0112] (4) The service asset intelligent classification module uses natural language processing technology to automatically classify service asset API interfaces;

[0113] (5) The feedback and optimization module collects user feedback.

[0114] As a preferred embodiment of the present invention, the step (1) specifically comprises the following steps:

[0115] (1.1) Define the service asset model and define data standard requirements for each field of the service asset model;

[0116] (1.2) Specify the target data source to be collected;

[0117] (1.3) According to the defined target data source, select the corresponding parser and parse the content in the target data source according to the defined service asset model;

[0118] (1.4) Define the collection frequency and collection mode, and automatically execute the collection task;

[0119] (1.5) Verify the format of the collected data to ensure its integrity and accuracy.

[0120] As a preferred embodiment of the present invention, the step (2) specifically comprises the following steps:

[0121] (2.1) Remove duplicate data from the data collected by the service asset data collection module;

[0122] (2.2) removing noise and irrelevant information from the data collected by the service asset data collection module;

[0123] (2.3) For each specific service asset API, obtain the call frequency from the log file or log center and perform activity statistics;

[0124] (2.4) All data collected by the service asset data collection module are unified into a standard data format;

[0125] (2.5) The data collected by the service asset data collection module shall be standardized according to the data standards required by the service asset model.

[0126] As a preferred embodiment of the present invention, the step (4) specifically comprises the following steps:

[0127] (4.1) Select some data from all the collected service asset API data as training sample sets, and classify and label these sample data. At the same time, set clear activity classification standards based on the call frequency of the service asset API interface, and determine the activity label of each service asset API interface;

[0128] (4.2) Encoding: Perform one-hot encoding conversion on the marked business classification labels and activity labels, and convert the classification labels into a numerical encoding form that can be processed by the machine learning model;

[0129] (4.3) Use deep learning framework to build classification algorithm model;

[0130] (4.4) Use the encoded training sample data to train the model. During the training process, the model performance is optimized by adjusting the model's hyperparameters. At the same time, the model's generalization ability is evaluated to prevent overfitting.

[0131] (4.5) Apply the trained model to the remaining service asset API interface data, perform model reasoning, and automatically classify new service asset API interfaces.

[0132] As a preferred embodiment of the present invention, the step (5) specifically comprises the following steps:

[0133] (5.1) Judge the model inference results and classify them into accurate and inaccurate categories;

[0134] (5.2) If it is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set, and the step (5.3) is continued; if it is judged to be accurate, the step is exited;

[0135] (5.3) Continue training the classification algorithm model.

[0136] The main feature of the device for realizing intelligent classification processing of enterprise service assets based on AIGC of the present invention is that the device comprises:

[0137] a processor configured to execute computer executable instructions;

[0138] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for realizing intelligent classification processing of enterprise service assets based on AIGC are implemented.

[0139] The processor for realizing intelligent classification and processing of enterprise service assets based on AIGC of the present invention has the main feature that the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the various steps of the above-mentioned method for realizing intelligent classification and processing of enterprise service assets based on AIGC are realized.

[0140] The computer-readable storage medium of the present invention has the main feature of storing a computer program thereon, and the computer program can be executed by a processor to implement the various steps of the above-mentioned method for realizing intelligent classification processing of enterprise service assets based on AIGC.

[0141] In a specific embodiment of the present invention, a system and method for realizing intelligent classification of enterprise service assets based on AIGC are disclosed, which can significantly improve the efficiency and accuracy of enterprise service asset classification, reduce manual intervention, and has important academic and practical value.

[0142] The present invention is directed to a system and method for realizing intelligent classification of enterprise service assets based on AIGC, and mainly achieves the following objectives:

[0143] 1. Intelligent collection of service assets: This system can support service asset managers to automatically and intelligently collect service assets from documents, codes, logs, etc., and automatically register services, realizing the intelligent collection and management of service assets, greatly improving the convenience of enterprise service asset management.

[0144] 2. Intelligent classification of service assets: By utilizing natural language processing technology, this system can automatically and intelligently classify service asset API interfaces, improving the accuracy and efficiency of classification, reducing the workload and error rate of manual classification, and making service asset management more refined.

[0145] 3. Intelligent activity classification of service assets: This system analyzes the frequency of API calls through log statistics and provides intelligent classification of activity to help enterprises identify key and commonly used service assets, thereby optimizing resource allocation and improving service efficiency.

[0146] 4. Business personnel participate in service asset optimization management in real time: Business personnel can provide feedback in real time, and the system conducts self-learning and optimization based on the feedback, so that business personnel can directly participate in the optimization management of service assets, improving business participation and the overall performance of the system. It not only reduces the burden on the service asset management team, but also promotes the digital transformation and innovation of enterprises.

[0147] The system of the present invention includes a service asset data acquisition module, a service asset data preprocessing module, a service asset registration module, a service asset intelligent classification module, and a feedback and optimization module.

[0148] The service asset data collection module automatically extracts service asset API interface information from code bases, documents and logs; the service asset data preprocessing module is responsible for cleaning, formatting and standardizing data; the service asset registration module registers the service interface collected by the service asset data collection module to the service asset library; the service asset intelligent classification module uses natural language processing technology to automatically classify service asset API interfaces; the feedback and optimization module collects user feedback to optimize system performance and accuracy.

[0149] The present invention is constituted as follows Figure 1 shown.

[0150] The service asset data collection module, service asset data preprocessing module, service asset registration module, service asset intelligent classification module, and feedback and optimization module will be described in detail below.

[0151] 1. Service asset data collection module

[0152] The service asset data collection module automatically extracts service asset API interface information from the specified code base, documents and logs according to the predefined service asset model. It can intelligently parse the target data source and execute the automated data collection process to ensure the efficient collection of service asset API information. In addition, the module also includes a data verification mechanism to ensure the integrity and accuracy of the collected service asset API data.

[0153] The steps of operation are as follows Figure 2 As shown:

[0154] Step 1: Define the service asset model: The service asset model includes: service name, service path, operation method, description, parameter list, request body, response type, response example, version, security, access restriction, document link, creation date, update date, and status; and define data standard requirements for each field of the service asset model. The data collected subsequently needs to be standardized according to the data standard requirements;

[0155] Step 2: Specify the target data source: Specify the target data source to be collected. The target data source supports code libraries (for example, GitLab, GitHub, Git, etc.), documents (World documents, Swagger documents, etc.), logs (for example, log files, search engine Elasticsearch, etc.), etc. This step can specify one target data source, and also supports multiple data sources;

[0156] Step 3: Parse the target data source: According to the target data source defined in "Step 2", select the corresponding parser and parse the content in the target data source according to the service asset model defined in "Step 1";

[0157] Step 4: Automatic data collection: define the collection frequency and collection mode (full, incremental), and automatically execute the collection task;

[0158] Step 5: Data Verification: Verify the format of the collected data to ensure the integrity and accuracy of the data.

[0159] 2. Service asset data preprocessing module

[0160] The service asset data preprocessing module is responsible for a series of data processing duties on the service asset API data collected by the "service asset data collection module", including data deduplication, data cleaning, data formatting, and data standardization. These steps are designed to improve the accuracy, consistency, and availability of data, thereby ensuring that the data quality reaches high standards and provides a reliable data foundation for subsequent service asset intelligent classification modules; at the same time, this module will count the activity of the service asset API for subsequent activity classification.

[0161] The steps of operation are as follows Figure 3 shown.

[0162] Step 1: Data deduplication: remove duplicate data from the data collected in the "Service Asset Data Collection Module";

[0163] Step 2: Data cleaning: Remove noise and irrelevant information from the data collected in the "Service Asset Data Collection Module", such as comments, blank lines, invalid spaces, etc.

[0164] Step 3: Activity statistics: For each specific service asset API, obtain the call frequency from the log file or log center, and support activity data in units of minutes, hours, days, months, and years;

[0165] Step 4: Data formatting: All data collected in the "Service Asset Data Collection Module" are unified into a standard data format: JSON format;

[0166] Step 5: Data standardization: The data collected in the "Service Asset Data Collection Module" is standardized according to the data standards required by the service asset model.

[0167] 3. Service asset registration module

[0168] The service asset registration module is responsible for receiving the output provided by the "service asset data collection module" and formally registering the collected service asset API interface information into the service asset library. To ensure accurate management and efficient retrieval, each registered service asset interface is assigned a unique identification ID. In addition, this module is also responsible for monitoring and updating the information in the service asset library to ensure that the records can accurately reflect the current status and historical changes of the service interface.

[0169] The main operations provided by the service asset registration module are:

[0170] 1) Register the service asset API to the service asset library based on the unique identification ID;

[0171] 2) Query the service asset API details based on the unique identification ID;

[0172] 3) Delete the service asset API based on the unique identification ID;

[0173] 4) Query the status information of the service asset API based on the unique identification ID;

[0174] 5) Query the classification information of the service asset API based on the unique identification ID;

[0175] 6) Update the service asset API based on the unique identification ID;

[0176] 7) Update the calling frequency of the service asset API based on the unique identification ID.

[0177] IV: Service asset intelligent classification module

[0178] The service asset intelligent classification module uses natural language processing technology to achieve automatic classification of service asset API interfaces. This module selects some data from the data collected by the service asset data collection module and the service asset data preprocessing module as training samples and performs manual classification and annotation to build a training sample set; when preparing training data, the business classification label and activity label are converted into one-hot encoding form for use in the training of deep learning models; the deep learning model is used to build a classification algorithm model, and the classification algorithm model is trained to enable it to intelligently classify new service asset APIs, thereby improving the accuracy and efficiency of classification.

[0179] The steps of operation are as follows Figure 4 shown.

[0180] Step 1: Training data annotation: Collect all service asset API data from the service asset data collection module and the service asset data preprocessing module, and select 10% (or other proportions) of the data as the training sample set. And classify and annotate these sample data, and clearly define the business classification label of each sample; at the same time, according to the call frequency of the service asset API interface, set clear activity classification standards and determine the activity label of each service asset API interface;

[0181] Step 2: Encoding: Perform one-hot encoding conversion on the business classification labels and activity labels marked in "Step 1" to convert the classification labels into a numerical encoding form that can be processed by the machine learning model to facilitate model input and processing;

[0182] Step 3: Build a classification algorithm model: Use a deep learning framework and build a convolutional neural network (CNN); the algorithm model includes an embedding layer for converting text into a numerical vector, followed by a combination of multiple convolutional layers and pooling layers for feature extraction and dimensionality reduction, and finally a fully connected layer for classification, and defines a business classification output layer and an activity classification output layer; this structure can effectively process the text description of the service asset API interface, extract key features, and use them for subsequent intelligent classification tasks;

[0183] Step 4: Train the classification algorithm model: Use the encoded training sample data prepared in "Step 2" to train the model; during the training process, optimize the performance of the model by adjusting the model's hyperparameters, such as learning rate, batch size, number of iterations, etc.; at the same time, apply techniques such as cross-validation to evaluate the generalization ability of the model and take measures to prevent overfitting;

[0184] Step 5: Model reasoning: Apply the trained model to the remaining service asset API interface data and perform model reasoning to automatically classify new service asset API interfaces; the model will output the predicted classification label and activity label for each interface to achieve intelligent classification of service asset APIs.

[0185] 5. Feedback and Optimization Module

[0186] The feedback and optimization module is designed to collect feedback from business personnel to continuously optimize system performance and accuracy. During the model inference phase, business personnel will manually review and judge based on the classification results of the model. If the classification results of the model are found to deviate from the actual situation, business personnel can mark these differences and feed them back to the system as new training data. This feedback data will be used to retrain and adjust the classification model to improve the accuracy and adaptability of the model. In this way, business personnel can not only participate in the intelligent classification process of the service asset API interface, but also directly contribute to the optimization of the model, forming a closed-loop learning and improvement mechanism.

[0187] The steps of operation are as follows Figure 5 as shown.

[0188] Step 1: Determine the model inference results: Determine the accuracy of the classification, which can be divided into accurate and inaccurate categories;

[0189] Step 2: Add to the training data sample set: If "Step 1" is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set; if "Step 1" is judged to be accurate, the data is ignored and the subsequent steps are ignored;

[0190] Step 3: Train the classification algorithm model: Continue to train the classification algorithm model.

[0191] In a specific embodiment of the present invention, this section takes the automatic collection of service asset API interfaces in the Swagger document, labeling them with corresponding business classification tags, and labeling them with activity tags according to the call frequency as an example to illustrate the implementation method of the present invention.

[0192] Next, according to the method of the present invention, a system and method for realizing intelligent classification of enterprise service assets based on AIGC is adopted. Through simple interaction, intelligent classification of enterprise service asset API interfaces is realized, and the convenience advantages of the present invention will be clearly felt.

[0193] 1. Service asset data collection module

[0194] This example automatically extracts service asset API interface information from the Swagger document. The target data source automatically collected in this example is the "SampleSwagger.yaml" file. The service asset API interface is automatically collected from the file and the data is verified.

[0195] The operation steps of the service asset data collection module are as follows:

[0196] 1. Define the service asset model

[0197] The service asset model includes fields such as service name, service path, operation method, description, parameter list, request body, response type, response example, version, security, access restriction, document link, creation date, update date, and status.

[0198]

[0199]

[0200]

[0201] Define the data standard of the data model. In this example, the "status" field is used as an example to provide the corresponding data standard. The data of "status" must come from the following enumeration values ​​to achieve the unification of the enterprise service asset model.

[0202] The data standard of the "Status" field is defined as follows:

[0203] ACTIVE / / Active status, indicating that the API interface is in normal use.

[0204] MAINTENANCE / / Maintenance status, indicating that the API interface is currently unavailable and is undergoing maintenance.

[0205] DEPRECATED / / Deprecated status, indicating that the API interface is outdated and is not recommended for use.

[0206] DISABLED / / Disabled state, indicating that the API interface has been disabled and no longer provides services.

[0207] ARCHIVED / / Archive status, indicating that the API interface has been archived and is inactive and unmaintained.

[0208] PLANNED / / Planning status, indicating that the API interface is planned for development but has not yet been implemented.

[0209] EXPERIMENTAL / / Experimental status, indicating that the API interface is in the experimental stage and may be unstable.

[0210] PUBLISHED / / Published status, indicating that the API interface has been developed and passed the review and is officially released to the public.

[0211] RETIRED / / Retired status, indicating that the API interface has been completely retired and is no longer supported.

[0212] The above example shows the service asset model definition and a sample of data standard requirements.

[0213] 2. Specify the target data source

[0214] Specify the target data source to be collected. This example obtains the service asset interface from the Swagger document and specifies the target data source as "SampleSwagger.yaml". The Swagger document defines three API interfaces: create order, get user list, and get product list.

[0215] The definition of SampleSwagger.yaml is as follows:

[0216]

[0217]

[0218]

[0219] 3. Analyze the target data source

[0220] Select the open source component Swagger Parser to parse the "SampleSwagger.yaml" file in "Step 2"; 4. Automatic data collection

[0221] Define the collection frequency as once a day, the collection window as "23:23" every day, and the collection mode as incremental collection;

[0222] The data collected from the three service asset APIs are as follows:

[0223]

[0224]

[0225] 5. Data Validation

[0226] The format of the data collected in "Step 4" is verified. The verification rules use the "Data Verification" defined in "Step 1" for verification. In this example, the "Status" field of the service asset API interface is verified, and the requirement is that it must be a "string". This example verifies whether the collected content is a string. The status of the three collected interfaces "Create Order", "Get User List", and "Get Product List" are all "ACTIVE", which meets the string requirement. If not, the user is prompted to revise the status field data;

[0227] 2. Service asset data preprocessing module

[0228] The service asset data preprocessing module is responsible for deduplication, clarification, formatting, standardization, and activity statistics of the service asset API interface data collected by the "service asset data collection module".

[0229] The operation steps of the service asset data preprocessing module are as follows:

[0230] 1. Data deduplication

[0231] Remove duplicate data from the data collected in the "Service Asset Data Collection Module" and determine whether there is duplicate data based on the four attributes of service name, service path, operation method, and version. Duplicate data needs to be removed. In this example, the three service asset API interfaces collected: "Create Order", "Get User List", and "Get Product List" do not contain duplicate data and do not need to be removed.

[0232] 2. Data cleaning

[0233] For the data collected in the "Service Asset Data Collection Module": "Create Order", "Get User List", "Get Product List", remove noise and irrelevant information in the data, such as comments, blank lines, invalid spaces, etc.;

[0234] The data cleaning sample code is as follows:

[0235] ```Python

[0236] def clean_api_data(api_list):

[0237] cleaned_apis = []

[0238] for api in api_list:

[0239] cleaned_api={key:api[key]for key in['path','operations']if key inapi}

[0240] cleaned_apis.append(cleaned_api)

[0241] return cleaned_apis

[0242] cleaned_api_list=clean_api_data(api_list)

[0243] 3. Activity Statistics

[0244] In this step, the call frequency of three service asset API interfaces, "Create Order", "Get User List", and "Get Product List" per hour is counted to provide hourly activity. The call frequency in the last hour is calculated from the log file: logfile.log.

[0245] The frequency statistics example code is as follows:

[0246] ```Python

[0247] import re

[0248] from datetime import datetime,timedelta

[0249] #The log file path is log_path

[0250] log_path="path / to / your / logfile.log"

[0251] #Read the log file and count the call frequency of each API interface

[0252] api_call_freq = {}

[0253] one_hour_ago=datetime.now()-timedelta(hours=1)

[0254] with open(log_path,'r')as file:

[0255] for line in file:

[0256]

[0257] 4. Data formatting

[0258] The data collected in the "Service Asset Data Collection Module": "Create Order", "Get User List", and "Get Product List" are uniformly formatted in JSON format. In this example, the data collected already exists in the form of a JSON file, so there is no need to format the data in this example. The following is a sample code for formatting data in JSON.

[0259] The data formatting sample code is as follows:

[0260]

[0261] 5. Data Standardization

[0262] In this example, the collected data: "Create Order", "Get User List", and "Get Product List" are standardized according to the data standard requirements of the service asset model defined in the "Service Asset Data Collection Module". This example uses the "Status Field" in the service asset API interface as an example. The data standard requires the "Status" field to be one of the following enumeration values: "ACTIVE, MAINTENANCE, DEPRECATED, PLANNED, EXPERIMENTAL, PUBLISHED, RETIRED". The three interfaces need to be checked and processed. If they do not belong to the enumeration values, the user is prompted to manually determine the status and correct the data to meet the data standards.

[0263] 3. Service asset registration module

[0264] The service asset registration module registers the data collected by the "service asset data collection module" into the service asset library, mainly using the service registration interface: registerServiceAssetAPI();

[0265] The main operation example interfaces provided by the service asset registration module are as follows:

[0266]

[0267]

[0268] IV: Service asset intelligent classification module

[0269] The service asset intelligent classification module uses natural language processing technology to achieve automatic classification of service asset API interfaces. In this example, some data is first selected from all data sets to be defined as training data and annotated, and the business classification and activity classification are encoded and processed. A neural network is used to build a classification algorithm model and algorithm training, and finally model reasoning is achieved to predict the intelligent classification of new service asset API interfaces.

[0270] The operation steps of the service asset intelligent classification module are as follows:

[0271] 1. Training data annotation

[0272] 10% of the data is selected from all service asset API interfaces, and training data annotation is done manually. In this example, the three service asset API interfaces collected: "Create Order", "Get User List", and "Get Product List" are all included in the training data, and the business categories are defined as "Transaction Class, Management Class, and Basic Class" respectively; define the call frequency setting of the service asset API interface. In this example, the activity classification standards are set as more than 100,000 times as high frequency, more than 50,000 and less than 100,000 times as medium frequency, and less than 50,000 times as low frequency, and the activity classification is defined as "high frequency, medium frequency, and low frequency" respectively;

[0273] The prepared training data is as follows:

[0274]

[0275]

[0276] 2. Encoding Processing

[0277] Encode the training data in step 1; in this example, the business categories are set as: management, transaction, monitoring, configuration, basic, and middle-end; the activity categories are: high frequency, medium frequency, and low frequency; then one-hot encoding conversion is performed on the above business categories and activity categories. This example uses the framework in the Python language to process and implement, converting the classification labels into numerical encoding forms that can be processed by the machine learning model;

[0278] Sample code for one-hot encoding conversion of business classification and activity classification:

[0279]

[0280]

[0281] The format of the training data after one-hot encoding is as follows:

[0282]

[0283] In the encoded training data, the one-hot encoding results of "Business Classification Label" and "Activity Label" are clearly displayed in their respective columns. For example, for the "Create Order" interface, the value of the "Transaction Class" column is 1, indicating that the interface belongs to the "Transaction Class", while the values ​​of other business classification label columns are 0. Similarly, the value of the "High Frequency" column is 1, indicating that the activity of the interface is "High Frequency";

[0284] 3. Build a classification algorithm model

[0285] This step uses Python and the deep learning framework Keras as an example to build a convolutional neural network. The algorithm model mainly includes an embedding layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The embedding layer converts text into a numerical vector, the convolution layer and the pooling layer extract key features, and the fully connected layer performs classification. Two output layers are added: one for multi-classification of business classification labels and the other for multi-classification of activity labels. Each output layer uses a "softmax" activation function to output the probability distribution of each category.

[0286] The classification algorithm model based on Python and deep learning framework Keras is as follows:

[0287]

[0288] The hyperparameters are, where “max_features” is the size of the vocabulary, “embed_size” is the dimension of the embedding vector, “input_length” is the maximum length of the input sequence, “num_business_categories” is the number of business category labels, and “num_activity_levels” is the number of activity level labels;

[0289] The embedding layer in the algorithm model is responsible for converting words in the text into numerical vectors. The convolution layer uses "ConvertD" to extract local features. The pooling layer uses "GlobalMaxPooling1D" for dimensionality reduction. The fully connected layer is used for business classification. The output layer is classified according to business labels and activity labels.

[0290] 4. Training classification algorithm model

[0291] This example always uses the sample data prepared in "Step 2" for training, and additionally prepares a test data set. The test set data is used to cross-validate the trained algorithm model to prevent the algorithm from overfitting.

[0292] The sample code implementation of the training algorithm is as follows:

[0293]

[0294]

[0295] In the data preparation stage, use "LabelEncoder" to convert business classification labels and activity labels into integer indexes, and use "to_categorical" to convert certificate indexes into one-hot encoding form;

[0296] Use "train_test_split" to divide the dataset into training and test sets;

[0297] Use "Sequential" to build the model defined in step 3 and add an output layer;

[0298] Compile the model using "adam" and specify different loss functions and evaluation metrics for each output layer;

[0299] Define callback functions, use "EarlyStopping" to stop training when the loss on the validation set no longer improves, and use "ModelCheckpoint" to save the best performing model on the validation set;

[0300] Train the model using "model.fit" and pass in the training data, labels, batch size, number of epochs, and validation data;

[0301] Use "model.evaluate" to evaluate the performance of the model on the test set;

[0302] Use “KFold” for cross validation to evaluate the generalization ability of the model;

[0303] Through the above training, the best training model obtained in this example is: "best_model.h5";

[0304] 5. Model Reasoning

[0305] In this example, a new service asset API data is provided, and the best model trained in step 4 is used for model inference to automatically classify the business and activity of the new service asset API data.

[0306] The following sample code demonstrates model inference:

[0307] ```Python

[0308] import numpy as np

[0309] import tensorflow as tf

[0310] from tensorflow.keras.models import load_model

[0311] from sklearn.preprocessing import LabelEncoder

[0312] model=load_model('best_model.h5')

[0313] #Preprocess the data to make it consistent with the training data format

[0314] def preprocess_data(data):

[0315] #Convert the data into the format required for model input

[0316] return data

[0317] #New service asset API data as test data

[0318] new_api_data={

[0319] "Service Name":"Query Order Status",

[0320] "Service path":" / orders / status",

[0321] "Operation method":"GET",

[0322] "Description":"Query the status of a specific order",

[0323] "Parameter list": [{"name":"orderId","type":"string","required":true}],

[0324] "Request body":{},

[0325] "Response Type":"application / j son",

[0326] "Response example":{"status":"pending"},

[0327] "Version":"1.0.0",

[0328] "security":"none",

[0329] "Access restrictions":{},

[0330] "Document Link": "https: / / testapi.example.com / docs / orders / status",

[0331] "Creation Date":"2024-11-04",

[0332] "Update Date":"2024-11-04",

[0333] "Status":"ACTIVE"

[0334] }

[0335] #Convert the new API data into the format required for model input

[0336] #preprocess_data function can process the entire dictionary and return an array suitable for the model

[0337] X_new_processed=preprocess_data([new_api_data])

[0338]

[0339] Use the "load_model" function to load the saved best model;

[0340] Use the new service asset API data "new_api_data" as test data;

[0341] Use the "model.predict" method to infer the preprocessed new data and obtain the predicted output of the model;

[0342] Use “LabelEncoder” to convert the predicted one-hot encoding back to the original label;

[0343] After the model inference is completed, the inference results are as follows:

[0344] New Service API-Predicted Business Category: Basic Class, Predicted Activity Level: Low Frequency

[0345] So far, this example has completed the service asset intelligent classification module scenario;

[0346] 5. Feedback and Optimization Module

[0347] In the feedback and optimization module, business personnel judge the correctness of the test data model reasoning results in the service asset intelligent classification module.

[0348] The operation steps of the feedback and optimization module are as follows:

[0349] 1. Determine the model inference results

[0350] Business personnel judge the correctness of the inference results of the test data "new_api_data";

[0351] 2. Add training data sample set

[0352] The business personnel determined that the test data "new_api_data" was incorrect, revised the business classification to "transaction type", and added the data to the sample set of training data;

[0353] 3. Training classification algorithm model

[0354] Continue to train the model using "Step 4: Train the classification algorithm model" in the Service Asset Intelligent Classification module;

[0355] According to the above steps, based on the target data source specified by the business personnel, the defined training data sample set, and the defined training classification algorithm model, the intelligent classification of service assets can be realized quickly and intelligently. The use of automated data collection, preprocessing, registration, and intelligent classification significantly improves the efficiency and accuracy of service asset management; the system uses natural language processing technology to realize automatic classification and activity analysis of API interfaces, reducing manual intervention and lowering the error rate; at the same time, the real-time feedback mechanism of business personnel enables the system to self-learn and optimize, enhance business participation, and improve system performance. The use of AIGC-based intelligent classification of enterprise service assets reduces the burden on the service asset management team, promotes the intelligence and automation of enterprise operations, and thus accelerates the widespread popularization and continuous innovation of digital applications.

[0356] The specific implementation scheme of this embodiment can refer to the relevant description in the above embodiment, which will not be repeated here.

[0357] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0358] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0359] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0360] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0361] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the corresponding program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.

[0362] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0363] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0364] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0365] The system, method, device, processor and computer-readable storage medium thereof for realizing intelligent classification and processing of enterprise service assets based on AIGC of the present invention have been adopted, and the efficiency and accuracy of service asset management have been significantly improved through automated data collection, preprocessing, registration and intelligent classification; the system uses natural language processing technology to realize automatic classification and activity analysis of service asset API interfaces, reducing manual intervention and lowering the error rate; at the same time, the real-time feedback mechanism of business personnel enables the system to self-learn and optimize, enhancing business participation and improving system performance. The intelligent classification of enterprise service assets based on AIGC has reduced the burden on the service asset management team, promoted the intelligence and automation of enterprise operations, and thus accelerated the widespread popularization and continuous innovation of digital applications.

[0366] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it is apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Therefore, the specification and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A system for realizing intelligent classification and processing of enterprise service assets based on AIGC, characterized in that: The system comprises: Service asset data collection module, used to automatically extract service asset API interface information from code bases, documents, and logs; A service asset data preprocessing module, connected to the service asset data collection module, is used to clean, format and standardize the collected service asset API data; A service asset registration module, connected to the service asset data preprocessing module, is used to receive the output of the service asset data collection module and formally register the collected service asset API interface information into the service asset library; A service asset intelligent classification module, connected to the service asset registration module, for automatically classifying service asset API interfaces using natural language processing technology; The feedback and optimization module is connected to the service asset intelligent classification module and is used to collect user feedback to optimize system performance and accuracy.

2. The system for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 1 is characterized in that: The service asset data collection module automatically extracts service asset API interface information from code bases, documents and logs, specifically including the following steps: (1.1) Define the service asset model and define data standard requirements for each field of the service asset model; (1.2) Specify the target data source to be collected; (1.3) According to the defined target data source, select the corresponding parser and parse the content in the target data source according to the defined service asset model; (1.4) Define the collection frequency and collection mode, and automatically execute the collection task; (1.5) Verify the format of the collected data to ensure its integrity and accuracy.

3. The system for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 1 is characterized in that: The service asset data preprocessing module cleans, formats and standardizes the collected service asset API data, specifically including the following steps: (2.1) Remove duplicate data from the data collected by the service asset data collection module; (2.2) removing noise and irrelevant information from the data collected by the service asset data collection module; (2.3) For each specific service asset API, obtain the call frequency from the log file or log center and perform activity statistics; (2.4) All data collected by the service asset data collection module are unified into a standard data format; (2.5) The data collected by the service asset data collection module shall be standardized according to the data standards required by the service asset model.

4. The system for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 1 is characterized in that: The service asset registration module assigns a unique identification ID to each registered service asset interface. The service asset registration module registers the service asset API to the service asset library, queries the service asset API details, deletes the service asset API, queries the status information of the service asset API, queries the classification information of the service asset API, updates the service asset API, and updates the calling frequency of the service asset API based on the unique identification ID.

5. The system for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 1 is characterized in that: The service asset intelligent classification module uses natural language processing technology to automatically classify service asset API interfaces, specifically including the following steps: (4.1) Select some data from all the collected service asset API data as training sample sets, and classify and label these sample data. At the same time, set clear activity classification standards based on the call frequency of the service asset API interface, and determine the activity label of each service asset API interface; (4.2) Encoding: Perform one-hot encoding conversion on the marked business classification labels and activity labels, and convert the classification labels into a numerical encoding form that can be processed by the machine learning model; (4.3) Use deep learning framework to build classification algorithm model; (4.4) Use the encoded training sample data to train the model. During the training process, the model performance is optimized by adjusting the model's hyperparameters. At the same time, the model's generalization ability is evaluated to prevent overfitting. (4.5) Apply the trained model to the remaining service asset API interface data, perform model reasoning, and automatically classify new service asset API interfaces.

6. The system for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 1 is characterized in that: The feedback and optimization module collects user feedback, specifically including the following steps: (5.1) Judge the model inference results and classify them into accurate and inaccurate categories; (5.2) If it is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set, and the step (5.3) is continued; if it is judged to be accurate, the step is exited; (5.3) Continue training the classification algorithm model.

7. A method for realizing intelligent classification and processing of enterprise service assets based on AIGC using the system of claim 1, characterized in that: The method comprises the following steps: (1) Service asset data collection module, which automatically extracts service asset API interface information from code bases, documents, and logs; (2) The service asset data preprocessing module cleans, formats, and standardizes the collected service asset API data; (3) The service asset registration module receives the output of the service asset data collection module and formally registers the collected service asset API interface information into the service asset library; (4) The service asset intelligent classification module uses natural language processing technology to automatically classify service asset API interfaces; (5) The feedback and optimization module collects user feedback.

8. The method for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 7 is characterized in that: The step (1) specifically comprises the following steps: (1.1) Define the service asset model and define data standard requirements for each field of the service asset model; (1.2) Specify the target data source to be collected; (1.3) According to the defined target data source, select the corresponding parser and parse the content in the target data source according to the defined service asset model; (1.4) Define the collection frequency and collection mode, and automatically execute the collection task; (1.5) Verify the format of the collected data to ensure its integrity and accuracy.

9. The method for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 7 is characterized in that: The step (2) specifically comprises the following steps: (2.1) Remove duplicate data from the data collected by the service asset data collection module; (2.2) removing noise and irrelevant information from the data collected by the service asset data collection module; (2.3) For each specific service asset API, obtain the call frequency from the log file or log center and perform activity statistics; (2.4) All data collected by the service asset data collection module are unified into a standard data format; (2.5) The data collected by the service asset data collection module shall be standardized according to the data standards required by the service asset model.

10. The method for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 7, characterized in that: The step (4) specifically comprises the following steps: (4.1) Select some data from all the collected service asset API data as training sample sets, and classify and label these sample data. At the same time, set clear activity classification standards based on the call frequency of the service asset API interface, and determine the activity label of each service asset API interface; (4.2) Encoding: Perform one-hot encoding conversion on the marked business classification labels and activity labels, and convert the classification labels into a numerical encoding form that can be processed by the machine learning model; (4.3) Use deep learning framework to build classification algorithm model; (4.4) Use the encoded training sample data to train the model. During the training process, the model performance is optimized by adjusting the model's hyperparameters. At the same time, the model's generalization ability is evaluated to prevent overfitting. (4.5) Apply the trained model to the remaining service asset API interface data, perform model reasoning, and automatically classify new service asset API interfaces.

11. The method for realizing intelligent classification and processing of enterprise service assets based on AIGC according to claim 7 is characterized in that: The step (5) specifically comprises the following steps: (5.1) Judge the model inference results and classify them into accurate and inaccurate categories; (5.2) If it is judged to be inaccurate, it is manually revised to an accurate business classification and added to the training data sample set, and the step (5.3) is continued; if it is judged to be accurate, the step is exited; (5.3) Continue training the classification algorithm model.

12. A device for implementing intelligent classification processing of enterprise service assets based on AIGC, characterized in that: The device comprises: a processor configured to execute computer executable instructions; A memory storing one or more computer executable instructions, wherein when the computer executable instructions are executed by the processor, each step of the method for realizing intelligent classification processing of enterprise service assets based on AIGC as described in any one of claims 7 to 11 is implemented.

13. A processor for implementing intelligent classification processing of enterprise service assets based on AIGC, characterized in that: The processor is configured to execute computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for realizing intelligent classification processing of enterprise service assets based on AIGC as described in any one of claims 7 to 11 are implemented.

14. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the method for realizing intelligent classification processing of enterprise service assets based on AIGC as described in any one of claims 7 to 11.

Images