Method and platform for automatically generating information management system based on artificial intelligence

Through the automated generation method of information management system based on artificial intelligence, the use of model inference services to generate code from natural language requirements and automatically test deployment, the problems of traditional development efficiency, high cost and difficult maintenance are solved, and the ability to quickly respond to business needs is achieved. .

CN119002927AInactive Publication Date: 2024-11-22BEIJING KUAFU TECHNOLOGY CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411220936.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

During the traditional software development process, the information management system has low development efficiency, long cycle, high cost, difficult maintenance, poor flexibility and adaptability, making it difficult to quickly respond to changes in business needs.

Method used

Adopt the automated generation method of information management system based on artificial intelligence, obtain standard software requirements documents from the natural language requirements input by users through model inference services, split technical development tasks, generate new codes and perform automatic testing and deployment, reducing manual coding and Testing process.

Benefits of technology

It lowers the technical threshold, accelerates development efficiency, reduces labor costs, improves the flexibility and adaptability of the system, and can quickly respond to changes in business needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119002927A_ABST
    Figure CN119002927A_ABST
Patent Text Reader

Abstract

The invention discloses an information management system automatic generation method and platform based on artificial intelligence, and the method comprises the steps: obtaining a natural language demand inputted by a user, clarifying the user demand through a model reasoning service, and generating a standard software demand document; identifying an interface document of the current application, and generating an interface document meeting the interface of the current application in combination with the standard software requirement document; according to the standard software requirement document and the interface document, the technology development task is split into sub-tasks, and sub-task documents and unit test cases are generated; understanding the subtask document and the current code through a model reasoning service, and generating a new code meeting the corresponding subtask; and after the new code is tested and repaired, automatic deployment of the application is carried out. Through the technical scheme of the invention, the technical threshold is reduced, the development efficiency is accelerated, the labor cost is reduced, and the business requirements which change continuously can be quickly responded.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of software design technology, and in particular to an information management system automatic generation method based on artificial intelligence and an information management system automatic generation platform based on artificial intelligence. Background Art

[0002] At present, in the traditional software development process, the development of information management systems usually requires multiple stages such as demand analysis, system design, coding implementation, testing and verification, deployment and maintenance, involving a lot of manual work and human intervention. Although some existing development tools and frameworks can improve development efficiency, the following problems still exist:

[0003] (1) Low development efficiency and long cycle: The manual coding and testing process is cumbersome, resulting in a long development cycle. From demand analysis to final product delivery, the entire process takes a long time.

[0004] (2) High cost: A large number of developers and testers are required, which increases labor costs.

[0005] (3) Difficult maintenance: As system functions increase and change, maintenance and upgrades become complex and difficult.

[0006] (4) Poor flexibility and adaptability: The system is unable to respond quickly to changing business needs. Summary of the invention

[0007] In response to the above problems, the present invention provides an artificial intelligence-based automatic generation method and platform for an information management system. Through the artificial intelligence model reasoning service, the natural language requirements input by the user are analyzed, the technical development tasks are disassembled, the code is automatically generated, and the deployment process is automatically tested. There is no need to write a large amount of code, so the corresponding information management system can be quickly and conveniently built, which lowers the technical threshold, speeds up development efficiency, and reduces labor costs, so that it can quickly respond to changing business needs.

[0008] To achieve the above object, the present invention provides an automatic generation method of an information management system based on artificial intelligence, comprising:

[0009] Obtain natural language requirements input by users, clarify user requirements through model reasoning services, and generate standard software requirement documents (Product Requirement Document, PRD);

[0010] Identify the interface document of the current application, and generate an interface document that meets the interface of the current application in combination with the standard software requirement document;

[0011] According to the standard software requirement document and the interface document, the technical development task is divided into subtasks, and subtask documents and unit test cases are generated;

[0012] Understanding the subtask document and the current code through a model reasoning service, and generating new code that satisfies the corresponding subtask;

[0013] After the new code is tested and repaired, the application is automatically deployed.

[0014] In the above technical solution, preferably, the specific process of clarifying user needs through the model reasoning service includes:

[0015] Through the model reasoning service, multiple rounds of question and answer sessions are conducted with users in the form of questions and answers, and reference answers are provided for users to answer quickly and accurately, in order to clarify the users' real needs.

[0016] In the above technical solution, preferably, the specific process of splitting the technical development task into subtasks and generating subtask documents and unit test cases includes:

[0017] Split technical development tasks into decoupled and independent subtasks;

[0018] Combine compilation and analysis tools to preprocess existing codes;

[0019] Redesign the class diagram and the sequence diagram by combining the standard software requirement document and the interface document through the model reasoning service;

[0020] According to the redesigned class diagram and sequence diagram, analyze and adjust the subtasks required for the development of business functions, and generate corresponding subtask documents and unit test cases;

[0021] The subtasks interact with each other through exposed APIs or method calls.

[0022] In the above technical solution, preferably, the specific process of testing and repairing the new code includes:

[0023] Use the compiler to fix compilation errors through model inference services;

[0024] Use the testing tool to fix test errors through model reasoning services;

[0025] Through model reasoning services, scanning tools are used to discover code quality issues and fix them;

[0026] Through the model inference service, test modules to work together and fix integration errors.

[0027] In the above technical solution, preferably, the model reasoning service is provided based on a language model and a code generation model, and the language model and the code generation model are obtained by fine-tuning training through GPT, LLaMa V2, Code llama and StarCoder.

[0028] The present invention also proposes an information management system automatic generation platform based on artificial intelligence, which uses the information management system automatic generation method based on artificial intelligence disclosed in any one of the above technical solutions, including:

[0029] The requirements analysis module is used to obtain the natural language requirements input by users, clarify the user requirements through model reasoning services, and generate standard software requirements documents;

[0030] An interface definition module is used to identify the interface document of the current application and generate an interface document that meets the interface of the current application in combination with the standard software requirement document;

[0031] A decomposition development module is used to split the technical development task into subtasks according to the standard software requirement document and the interface document, and generate subtask documents and unit test cases;

[0032] A code generation module, used for understanding the subtask document and the current code through a model reasoning service, and generating new code that satisfies the corresponding subtask;

[0033] The test deployment module is used to automatically deploy the application after testing and repairing the new code.

[0034] In the above technical solution, preferably, the demand analysis module is specifically used to:

[0035] Through the model reasoning service, multiple rounds of question and answer sessions are conducted with users in the form of questions and answers, and reference answers are provided for users to answer quickly and accurately, in order to clarify the users' real needs.

[0036] In the above technical solution, preferably, the decomposition and development module is specifically used for:

[0037] Split technical development tasks into decoupled and independent subtasks;

[0038] Combine compilation and analysis tools to preprocess existing codes;

[0039] Redesign the class diagram and the sequence diagram by combining the standard software requirement document and the interface document through the model reasoning service;

[0040] According to the redesigned class diagram and sequence diagram, analyze and adjust the subtasks required for the development of business functions, and generate corresponding subtask documents and unit test cases;

[0041] The subtasks interact with each other through exposed APIs or method calls.

[0042] In the above technical solution, preferably, the test deployment module is specifically used to:

[0043] Use the compiler to fix compilation errors through model inference services;

[0044] Use the testing tool to fix test errors through model reasoning services;

[0045] Through model reasoning services, scanning tools are used to discover code quality issues and fix them;

[0046] Through the model inference service, test modules to work together and fix integration errors.

[0047] In the above technical solution, preferably, the model reasoning service is provided based on a language model and a code generation model, and the language model and the code generation model are obtained by fine-tuning training through GPT, LLaMa V2, Code llama and StarCoder.

[0048] Compared with the prior art, the beneficial effects of the present invention are: through the model reasoning service of artificial intelligence, the natural language requirements input by the user are analyzed, the technical development tasks are disassembled, the code is automatically generated and the automatic test deployment process is carried out. There is no need to write a large amount of code, which makes it convenient to quickly build the corresponding information management system, lowers the technical threshold, speeds up development efficiency, reduces labor costs, and enables it to quickly respond to changing business needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flowchart of an automatic generation method of an information management system based on artificial intelligence disclosed in an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of a model training process disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0053] like Figure 1 As shown, according to an artificial intelligence-based automatic generation method of an information management system provided by the present invention, the method comprises:

[0054] Obtain natural language requirements input by users, clarify user requirements through model reasoning services, and generate standard software requirements documents;

[0055] Identify the interface document of the current application, combine it with the standard software requirement document, and generate an interface document that meets the current application interface;

[0056] According to standard software requirement documents and interface documents, split technical development tasks into subtasks, and generate subtask documents and unit test cases;

[0057] The model inference service understands the subtask documents and current code, and generates new code that satisfies the corresponding subtasks.

[0058] After testing and fixing the new code, the application is automatically deployed.

[0059] In this implementation, through the artificial intelligence model reasoning service, the natural language requirements input by the user are analyzed, the technical development tasks are broken down, the code is automatically generated, and the automatic test deployment process is carried out. There is no need to write a large amount of code, which makes it convenient to quickly build a corresponding information management system, lowers the technical threshold, speeds up development efficiency, and reduces labor costs, enabling it to quickly respond to changing business needs.

[0060] Specifically, the information management systems generated by this method include book management systems, store management systems, project management systems, business opportunity management systems, production line management systems, and production process management systems.

[0061] During the implementation process, users input their requirements in natural language. Based on the obtained natural language requirements, natural language processing technology is used to understand the requirements, and intelligent analysis is performed on the requirements content to generate standardized software requirements documents. In the process of requirements analysis, it is necessary to consider the knowledge of existing services and clarify the requirements.

[0062] Generate intelligent interfaces based on requirements, including standard software requirement documents, interface definitions, etc. The interfaces cover the calling interfaces between front-end and back-end interfaces and services.

[0063] The generation process includes: real-time identification of the current application's swagger, RPC, GraphQL, API Blueprint and other interface documents, combined with standard software requirement documents, designing and adjusting interfaces through model reasoning services, and generating interface documents that meet the requirements.

[0064] When the interface is adapted, since a development task may involve modifying multiple files, the development task is broken down into multiple development subtasks, such as methods or business logic under a certain file of a software project.

[0065] Based on the subtask document, the corresponding new code is generated for the content of the subtask and the current existing code to meet the requirements of the subtask. Preferably, the code needs to be checked to ensure that there are no problems and the subtask can be completed.

[0066] By integrating continuous integration CI tools, the collaborative work of each subtask module is systematically tested and automatically repaired, and docking with major cloud vendors (Alibaba Cloud, Huawei Cloud, Tencent Cloud, AWS, etc.), k8s and containerized platforms is achieved to complete the automated deployment of the corresponding information management system.

[0067] In the above implementation, preferably, the specific process of clarifying user requirements through the model reasoning service includes:

[0068] Through the model reasoning service, multiple rounds of question and answer sessions are conducted with users in the form of questions and answers, and reference answers are provided for users to answer quickly and accurately, in order to clarify the users' real needs.

[0069] In the above implementation, preferably, the specific process of splitting the technical development task into subtasks and generating subtask documents and unit test cases includes:

[0070] Split technical development tasks into decoupled and independent subtasks;

[0071] Combine compilers, AST analysis and other compilation analysis tools to preprocess existing codes;

[0072] Redesign class diagrams and sequence diagrams by combining model reasoning services with standard software requirement documents and interface documents;

[0073] According to the redesigned class diagram and sequence diagram, analyze and adjust the subtasks required for the development of business functions, and generate corresponding subtask documents and unit test cases;

[0074] Among them, subtasks interact with each other through exposed APIs or method calls.

[0075] In the above implementation, preferably, the specific process of testing and repairing the new code includes:

[0076] Fix compilation errors using compilers such as Maven through model inference services;

[0077] Fix test errors using Junit and other testing tools through model inference services;

[0078] Through the model reasoning service, use scanning tools such as Lint and CheckStyle to discover code quality problems and fix them;

[0079] Through the model inference service, test modules to work together and fix integration errors.

[0080] like Figure 2 As shown, in the above implementation, preferably, the model reasoning service is provided based on the language model and the code generation model, and the language model and the code generation model are obtained by fine-tuning and training through GPT, LLaMa V2, Code llama and StarCoder.

[0081] The present invention further proposes an information management system automatic generation platform based on artificial intelligence, which uses the information management system automatic generation method based on artificial intelligence disclosed in any one of the above embodiments, including:

[0082] The requirements analysis module is used to obtain the natural language requirements input by users, clarify the user requirements through model reasoning services, and generate standard software requirements documents;

[0083] The interface definition module is used to identify the interface document of the current application and generate an interface document that meets the interface of the current application in combination with the standard software requirement document;

[0084] Decomposition development module, which is used to split technical development tasks into subtasks according to standard software requirement documents and interface documents, and generate subtask documents and unit test cases;

[0085] The code generation module is used to understand the subtask documents and current code through the model reasoning service and generate new code that satisfies the corresponding subtask;

[0086] The test deployment module is used to automatically deploy applications after testing and repairing new codes.

[0087] In the above implementation, preferably, the demand analysis module is specifically used for:

[0088] Through the model reasoning service, multiple rounds of question and answer sessions are conducted with users in the form of questions and answers, and reference answers are provided for users to answer quickly and accurately, in order to clarify the users' real needs.

[0089] In the above implementation, preferably, the decomposition development module is specifically used for:

[0090] Split technical development tasks into decoupled and independent subtasks;

[0091] Combine compilation and analysis tools to preprocess existing codes;

[0092] Redesign class diagrams and sequence diagrams by combining model reasoning services with standard software requirement documents and interface documents;

[0093] According to the redesigned class diagram and sequence diagram, analyze and adjust the subtasks required for the development of business functions, and generate corresponding subtask documents and unit test cases;

[0094] Among them, subtasks interact with each other through exposed APIs or method calls.

[0095] In the above implementation, preferably, the test deployment module is specifically used to:

[0096] Use the compiler to fix compilation errors through model inference services;

[0097] Use the testing tool to fix test errors through model reasoning services;

[0098] Through model reasoning services, scanning tools are used to discover code quality issues and fix them;

[0099] Through the model inference service, test modules to work together and fix integration errors.

[0100] In the above implementation, preferably, the model reasoning service is provided based on the language model and the code generation model, and the language model and the code generation model are obtained by fine-tuning and training through GPT, LLaMa V2, Code llama and StarCoder.

[0101] According to the artificial intelligence-based information management system automatic generation platform disclosed in the above-mentioned implementation mode, the functions to be implemented by each module thereof correspond to the steps of the artificial intelligence-based information management system automatic generation method disclosed in the above-mentioned implementation mode. During the implementation process, operations are performed with reference to the above-mentioned implementation mode, which will not be repeated here.

[0102] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An automatic generation method of information management system based on artificial intelligence, It is characterized in that include: Obtain natural language requirements input by users, clarify user requirements through model reasoning services, and generate standard software requirements documents; Identify the interface document of the current application, and generate an interface document that meets the interface of the current application in combination with the standard software requirement document; According to the standard software requirement document and the interface document, the technical development task is divided into subtasks, and subtask documents and unit test cases are generated; Understanding the subtask document and the current code through a model reasoning service, and generating new code that satisfies the corresponding subtask; After the new code is tested and repaired, the application is automatically deployed.

2. The method for automatically generating an information management system based on artificial intelligence according to claim 1, It is characterized in that The specific process of clarifying user requirements through the model reasoning service includes: Through the model reasoning service, multiple rounds of question and answer sessions are conducted with users in the form of questions and answers, and reference answers are provided for users to answer quickly and accurately, in order to clarify the users' real needs.

3. The method for automatically generating an information management system based on artificial intelligence according to claim 1, It is characterized in that The specific process of splitting the technical development task into subtasks and generating subtask documents and unit test cases includes: Split technical development tasks into decoupled and independent subtasks; Combine compilation and analysis tools to preprocess existing codes; Redesign the class diagram and the sequence diagram by combining the standard software requirement document and the interface document through the model reasoning service; According to the redesigned class diagram and sequence diagram, analyze and adjust the subtasks required for the development of business functions, and generate corresponding subtask documents and unit test cases; The subtasks interact with each other through exposed APIs or method calls.

4. The method for automatically generating an information management system based on artificial intelligence according to claim 1, It is characterized in that The specific process of testing and repairing the new code includes: Use the compiler to fix compilation errors through model inference services; Use the testing tool to fix test errors through model reasoning services; Through model reasoning services, scanning tools are used to discover code quality issues and fix them; Through the model inference service, test modules to work together and fix integration errors.

5. The method for automatically generating an information management system based on artificial intelligence according to any one of claims 1 to 4, It is characterized in that The model reasoning service is provided based on a language model and a code generation model, and the language model and the code generation model are obtained by fine-tuning and training with GPT, LLaMa V2, Code llama and StarCoder.

6. An information management system automatic generation platform based on artificial intelligence, It is characterized in that The method for automatically generating an information management system based on artificial intelligence as claimed in any one of claims 1 to 5 comprises: The requirements analysis module is used to obtain the natural language requirements input by users, clarify the user requirements through model reasoning services, and generate standard software requirements documents; An interface definition module is used to identify the interface document of the current application and generate an interface document that meets the interface of the current application in combination with the standard software requirement document; A decomposition development module is used to split the technical development task into subtasks according to the standard software requirement document and the interface document, and generate subtask documents and unit test cases; A code generation module, used for understanding the subtask document and the current code through a model reasoning service, and generating new code that satisfies the corresponding subtask; The test deployment module is used to automatically deploy the application after testing and repairing the new code.

7. The information management system automatic generation platform based on artificial intelligence according to claim 6, It is characterized in that The demand analysis module is specifically used for: Through the model reasoning service, multiple rounds of question and answer sessions are conducted with users in the form of questions and answers, and reference answers are provided for users to answer quickly and accurately, in order to clarify the users' real needs.

8. The information management system automatic generation platform based on artificial intelligence according to claim 6, It is characterized in that The decomposition development module is specifically used for: Split technical development tasks into decoupled and independent subtasks; Combine compilation and analysis tools to preprocess existing codes; Redesign the class diagram and the sequence diagram by combining the standard software requirement document and the interface document through the model reasoning service; According to the redesigned class diagram and sequence diagram, analyze and adjust the subtasks required for the development of business functions, and generate corresponding subtask documents and unit test cases; The subtasks interact with each other through exposed APIs or method calls.

9. The information management system automatic generation platform based on artificial intelligence according to claim 6, It is characterized in that The test deployment module is specifically used for: Use the compiler to fix compilation errors through model inference services; Use the testing tool to fix test errors through model reasoning services; Through model reasoning services, scanning tools are used to discover code quality issues and fix them; Through the model inference service, test modules to work together and fix integration errors.

10. The information management system automatic generation platform based on artificial intelligence according to claim 6, It is characterized in that The model reasoning service is provided based on a language model and a code generation model, and the language model and the code generation model are obtained by fine-tuning and training with GPT, LLaMa V2, Code llama and StarCoder.

Citation Information

Cited By

  • SVG generation system and method based on large language model

    CN119668582A

  • Geographic knowledge complex question and answer method based on space-time agent architecture

    CN120745840A