Software development method, system and readable storage medium based on natural language programming

Abstract

This invention relates to a software development method, system, and readable storage medium based on natural language orchestration. The method includes: deploying a programmable system based on natural language orchestration; registering basic resources in the programmable system; describing the registered resources and system resources in natural language and storing them in a relational database and synchronizing them to a vector database; creating and orchestrating business applications in the programmable system; describing the orchestrated content in natural language and publishing it as an agent, storing it in the relational database and synchronizing it to the vector database; using a large language model to perform vector retrieval on the user's query content and selecting the most suitable agent; calling the most suitable agent to execute the corresponding orchestrated content in paragraph order and outputting the final result. The software development method provided by this invention allows users to break down requirements into detailed steps using natural language without needing to understand the technical principles, which are then executed by a large language model, reducing communication time and the possibility of requirement deviations.

Description

Technical Field

[0001] This invention relates to the field of software development technology, and in particular to a software development method, system, and readable storage medium based on natural language orchestration. Background Technology

[0002] With the acceleration of digital transformation, enterprises have an increasingly urgent need for automated and intelligent workflow management. In the process of software development and business system implementation, how to efficiently transform complex human instructions into executable program instructions to meet the needs of various software development businesses has always been a core problem that urgently needs to be solved. Currently, mainstream implementation methods have the following significant drawbacks:

[0003] 1. Inefficiency of traditional software development models

[0004] In traditional software engineering practices (such as development using programming languages ​​like Java and Python), the implementation of business requirements heavily relies on specialized developers. The complete cycle from requirements analysis to code writing, testing, and deployment often takes weeks or even months, involving numerous cross-departmental requirements communication meetings and repeated revisions. Even with detailed requirements documents, developers still need to translate the logic described in natural language layer by layer into a computer-readable code structure. This model not only leads to high labor costs but also requires recoding and redeploying the system when requirements change, resulting in a severe lack of flexibility and responsiveness.

[0005] 2. Limitations of Low-Code Platforms

[0006] In recent years, low-code / no-code development platforms (such as OutSystems and Bubble) have emerged, significantly lowering the development threshold through visual component drag-and-drop and pre-defined logic rules. However, these platforms still require users to have basic technical knowledge, and their primary target users remain technically savvy, requiring ongoing communication and collaboration with stakeholders with business needs. Summary of the Invention

[0007] The purpose of this invention is to provide a software development method, system, and readable storage medium based on natural language arrangement, so as to solve the problems of high development threshold, low efficiency, and high cost of mainstream software development methods.

[0008] To achieve the above objectives, this invention provides a software development method based on natural language orchestration, comprising the following steps:

[0009] S1. Deploy a programmable system based on natural language orchestration, register basic resources in the programmable system, describe the registered resources and system resources in natural language, store them in a relational database, and synchronize them to a vector database.

[0010] S2. Create and orchestrate business applications in the programmable system, then describe the orchestration content in natural language, publish it as an intelligent agent, store it in the relational database, and synchronize it to the vector database.

[0011] S3. Use a large language model to perform vector retrieval on the user's question content and select the most suitable agent;

[0012] S4. Call the most suitable agent to execute the corresponding arrangement content in paragraph order and output the final result.

[0013] Optionally, in S1, the registration of the basic resources includes interface resource registration and database resource publishing.

[0014] Optionally, in S1, the registered resources and system resources after the natural language description are first converted into SQL statements, and then the SQL statements are published as a standard RESTful interface before being stored in the relational database and synchronized to the vector database.

[0015] Optionally, step S3 specifically includes:

[0016] The large language model is used to perform vector retrieval on the question content to identify multiple agents;

[0017] The multiple agents are placed into the parameter block of the Function calling module of the large language model, and the large language model selects the most suitable agent.

[0018] Optionally, step S4 specifically includes:

[0019] S41. Call the most suitable intelligent agent to perform vector retrieval on the content of each paragraph in the order of the paragraphs, and find multiple interface resources;

[0020] S42. Place the multiple interface resources into the parameter block of the Function calling module, and the large language model selects the most suitable interface resource.

[0021] S43. Extract parameters from the question content and the returned content of the preceding paragraph, and use the extracted parameters to call the most suitable interface resource;

[0022] S44. Repeat S41-S43 until all paragraphs have been executed;

[0023] S45. Output the final result.

[0024] Based on the same inventive concept, the present invention also provides a programmable system based on natural language arrangement, comprising:

[0025] The data management module is used for registering basic resources;

[0026] The resource management module is used to describe registered resources and system resources in natural language, store them in a relational database, and synchronize them to a vector database.

[0027] Human-computer interaction interface, used to interact with users and create business applications;

[0028] The orchestration module is used to orchestrate the business application, and then describe the orchestration content in natural language, publish it as an intelligent agent, store it in the relational database, and synchronize it to the vector database.

[0029] The large language model is used to perform vector retrieval on the user's query content, select the most suitable agent, and call the most suitable agent to execute the corresponding arranged content in paragraph order, and output the final result.

[0030] Optionally, the programmable system further includes a text2sql module, which is used to convert the registered resources and system resources after natural language description into SQL statements, publish the SQL statements as a standard RESTful interface, store them in the relational database, and synchronize them to the vector database.

[0031] Optionally, the large language model includes a function calling module, which is specifically used for:

[0032] Vector retrieval is performed on the query content to find multiple agents, and these agents are placed into the parameter block of the Function calling module to select the most suitable agent for invocation and execution.

[0033] Optionally, the large language model is further used for:

[0034] The most suitable agent is invoked to perform vector retrieval on the content of each paragraph in the order of the paragraphs, and multiple interface resources are found.

[0035] The multiple interface resources are placed into the parameter block of the Function calling module to select the most suitable interface resource;

[0036] The parameters are extracted from the question content and the returned content of the preceding paragraphs, and the extracted parameters are used to call the most suitable interface resource until all paragraphs are executed. Then the final result is output to the human-computer interaction interface.

[0037] Based on the same inventive concept, the present invention also provides a readable storage medium having a computer program stored thereon, which, when executed, enables the software development method based on natural language arrangement as described above.

[0038] In the software development method, system, and readable storage medium based on natural language orchestration provided by this invention, users do not need to understand the technical principles. They can break down business requirements into detailed steps using natural language, which are then executed by a large language model. This method realizes natural language orchestration, which can lower the threshold, enable users to complete software development on their own, reduce communication time, and reduce the possibility of requirement deviations. Attached Figure Description

[0039] Those skilled in the art will understand that the accompanying drawings are provided to better understand the invention and do not constitute any limitation on the scope of the invention. Wherein:

[0040] Figure 1 The overall flowchart of a software development method based on natural language orchestration provided in an embodiment of the present invention;

[0041] Figure 2 A flowchart illustrating the application creation process provided in an embodiment of the present invention;

[0042] Figure 3 This is an application call flowchart provided in an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, advantages, and features of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the drawings are in a very simplified form and use non-precise proportions, and are only used to facilitate and clearly illustrate the purpose of the embodiments of this invention. Please refer to the accompanying drawings to make the objectives, features, and advantages of this invention more apparent and understandable. It should be understood that the structures, proportions, sizes, etc., depicted in the accompanying drawings are only used to complement the content disclosed in the specification, for those skilled in the art to understand and read, and are not intended to limit the implementation conditions of this invention. Any modifications to the structure, changes in proportions, or adjustments to the size, if they are the same as or similar to the effects and objectives achieved by this invention, should still fall within the scope of the technical content disclosed in this invention.

[0044] As used herein, the singular forms “a,” “an,” and “the” include plural objects unless otherwise expressly indicated. As used herein, the term “or” is generally used to include the meaning of “and / or” unless otherwise expressly indicated.

[0045] Please refer to Figures 1-3This embodiment provides a software development method based on natural language orchestration, including the following steps:

[0046] S1. Deploy a programmable system based on natural language orchestration, register basic resources in the programmable system, describe the registered resources and system resources in natural language, store them in a relational database, and synchronize them to a vector database.

[0047] S2. Create and orchestrate business applications in the programmable system, then describe the orchestration content in natural language, publish it as an intelligent agent, store it in the relational database, and synchronize it to the vector database.

[0048] S3. Use a large language model to perform vector retrieval on the user's question content and select the most suitable agent;

[0049] S4. Call the most suitable agent to execute the corresponding arrangement content in paragraph order and output the final result.

[0050] It should be noted that natural language orchestration refers to describing specific work content through language, textualizing human communication and guidance methods, and breaking down complex tasks in a manner as detailed as the step-by-step descriptions in various standardized operating manuals. Before the application of the software development method based on natural language orchestration provided by this invention, software development typically required professional programmers to implement it, and the process from requirement formulation to communication to coding and deployment required a significant amount of time and effort.

[0051] From a macro perspective, software products are all about data processing and presentation. To lower the development threshold and reduce communication time, enabling various users to quickly develop application services, this invention provides a software development method based on natural language orchestration. This method allows users to break down business requirements into detailed steps using natural language, without needing to understand technical principles, and then have them executed by a large language model. This method achieves natural language orchestration, lowering the barrier to entry, allowing users to complete software development independently, reducing communication time, and minimizing the possibility of requirement discrepancies.

[0052] First, execute step S1 to deploy a programmable system based on natural language orchestration, register basic resources in the programmable system, and store the registered resources and system resources in a relational database and synchronize them to the vector database after describing them in natural language.

[0053] In this embodiment, the registration of basic resources includes interface resource registration and database resource publishing. For example, the required standard RESTful interface information and data tables and fields from different databases can be described in natural language, stored in the relational database, and synchronized to the vector database.

[0054] In this embodiment, the system resources are the system's built-in basic resources, such as a knowledge base. This knowledge base can support the input of documents and question-answer pairs and vector retrieval, which can be used by users for arrangement.

[0055] Preferably, after describing the registered resources and system resources in natural language, and before storing them in the relational database and synchronizing them to the vector database, the registered resources and system resources described in natural language are first converted into SQL statements. Then, the SQL statements are published as a standard RESTful interface before being stored in the relational database and synchronized to the vector database, so that they can be called by the large language model later.

[0056] Then, S2 is executed to create and orchestrate a business application in the programmable system. The orchestrated content is then described in natural language, published as an agent, stored in the relational database, and synchronized to the vector database.

[0057] In this embodiment, orchestration can be managed at the level of "intelligent agents." After an intelligent agent is created, business content is orchestrated under that agent. During orchestration, business steps are described line by line, and each step must have clear logic, that is, clearly specifying the resources to be called or clearly describing the data processing requirements. After orchestration is completed, the intelligent agent is published to the market, and front-end users can call the intelligent agent through human-computer interaction.

[0058] Next, execute S3, such as... Figure 3 As shown, a large language model is used to perform vector retrieval on the user's query content to select the most suitable agent. In this embodiment, step S3 specifically includes:

[0059] The large language model is used to perform vector retrieval on the question content to identify multiple agents;

[0060] The multiple agents are placed into the parameter block of the Function calling module of the large language model, and the large language model selects the most suitable agent.

[0061] The Function Calling module is used by the large language model to convert text commands output by the model into specific function calls during interaction. This module is configured as a JSON string, primarily in the `tools` property, which includes the function name, description, parameters, and prompts. When the large language model interacts with the user, it identifies the intent in the user's question and selects the appropriate function to call and execute.

[0062] In this embodiment, function calls are implemented using vector retrieval. During user interaction, resource descriptions and call information are first matched in the vector library. Then, the matched description information is combined into the parameter block content of the function calling module, which is then executed by the large language model. This configuration is more accurate than directly placing all resources into the parameter block of the function calling module, reducing the number of parameters and improving the model's recognition performance and response speed.

[0063] Finally, S4 is executed, calling the most suitable agent to execute the corresponding arrangement content according to paragraph order, and outputting the final result. In this embodiment, please continue to refer to... Figure 3 The steps in S4 specifically include:

[0064] S41. Call the most suitable intelligent agent to perform vector retrieval on the content of each paragraph in the order of the paragraphs, and find multiple interface resources;

[0065] S42. Place the multiple interface resources into the parameter block of the Function calling module, and the large language model selects the most suitable interface resource.

[0066] S43. Extract parameters from the question content and the returned content of the preceding paragraph, and use the extracted parameters to call the most suitable interface resource;

[0067] S44. Repeat S41-S43 until all paragraphs have been executed;

[0068] S45. Output the final result.

[0069] In some specific embodiments, during paragraph execution, the function calling module is used to match the user's query content against the vector library to query potentially used interfaces. To ensure accuracy, each paragraph description matches at least 5 data entries in the vector library, without any similarity limit. Then, the matched RESTful interfaces from the vector library are placed into the parameter block of the function-calling module, and the large language model selects the most suitable interface resource.

[0070] The same approach is used when executing the next segment, but the execution result of the previous segment is added to the dialog context. S41-S43 are repeated until all segments are executed, and the execution result of the last segment is returned. This method allows for the assembly of services using natural language, and can be extended to include other types of services in the future.

[0071] Based on the same inventive concept, embodiments of the present invention also propose a programmable system based on natural language arrangement, comprising:

[0072] The data management module is used for registering basic resources;

[0073] The resource management module is used to describe registered resources and system resources in natural language, store them in a relational database, and synchronize them to a vector database.

[0074] Human-computer interaction interface, used to interact with users and create business applications;

[0075] The orchestration module is used to orchestrate the business application, and then describe the orchestration content in natural language, publish it as an intelligent agent, store it in the relational database, and synchronize it to the vector database.

[0076] The large language model is used to perform vector retrieval on the user's query content, select the most suitable agent, and call the most suitable agent to execute the corresponding arranged content in paragraph order, and output the final result.

[0077] In this embodiment, the data management module can implement different code for different databases based on J2EE interfaces and abstract concepts to support access to multiple databases. Then, a data dictionary is used to describe the tables and fields, completing the preliminary preparations for data querying.

[0078] In this embodiment, the registration of basic resources includes interface resource registration and database resource publishing. The system resources are basic resources that come with the system, such as a knowledge base. The resource management module manages all accessed resources, text2sql resources and system resources. Accessed resources are stored in the vector library after being described in natural language, and information is provided through vector retrieval service when used.

[0079] In this embodiment, the human-computer interaction interface can adopt a dialogue interaction mode, and different results can be rendered on the interaction interface, such as displaying statistical data in a table or bar chart after querying statistical data in the database.

[0080] In this embodiment, the orchestration module can manage by "intelligent agent". After creating an intelligent agent, business content is orchestrated under that intelligent agent. During orchestration, business steps are described line by line, and each step must have clear logic, that is, clearly define the resources to be called or clearly describe the data processing requirements. After orchestration is completed, the intelligent agent is published to the market, and front-end users can call the intelligent agent through the human-computer interaction interface.

[0081] Preferably, the programmable system further includes a text2sql module, which is used to convert the registered resources and system resources after natural language description into SQL statements, publish the SQL statements as a standard RESTful interface, store them in the relational database, and synchronize them to the vector database.

[0082] In this embodiment, the Text2sql module accesses the data source and data dictionary of the data management module. Users can submit data query requests through dialogue. After the large language model performs inference analysis, it generates SQL statements based on the data dictionary, which are then executed by the data management module.

[0083] When the text2sql module is invoked, the user's query is retrieved from the data table vector library to determine the approximate range of the user's tables. Then, based on the user's query, table and field information is retrieved from the data field vector library. Leveraging the reasoning capabilities of the large language model, all retrieved table and field information is placed into the context, and the large language model generates an SQL statement. This generated SQL statement is then executed in the corresponding data source, and the result is returned. After confirming the SQL execution is successful, a standard RESTful API can be published using this SQL.

[0084] This embodiment combines the text2sql module with vector retrieval technology, storing table information and field information separately in a vector library. When in use, the user dialogue is first matched with the table information vector, and then the dialogue, table name, and field information are matched with vectors to obtain a set of table and field descriptions that better meet the user's needs.

[0085] In this embodiment, the large language model includes a function calling module, and the large language model is specifically used for:

[0086] Vector retrieval is performed on the query content to find multiple agents, and these agents are placed into the parameter block of the Function calling module to select the most suitable agent for invocation and execution.

[0087] In this embodiment, the large language model is further used for:

[0088] The most suitable agent is invoked to perform vector retrieval on the content of each paragraph in the order of the paragraphs, and multiple interface resources are found.

[0089] The multiple interface resources are placed into the parameter block of the Function calling module to select the most suitable interface resource;

[0090] The parameters are extracted from the question content and the returned content of the preceding paragraphs, and the extracted parameters are used to call the most suitable interface resource until all paragraphs are executed. Then the final result is output to the human-computer interaction interface.

[0091] In some specific embodiments, during paragraph execution, a function calling module is used to match the user's query content against the vector library, querying for potentially useful interfaces. To ensure accuracy, each paragraph description matches at least 5 data entries in the vector library, without any similarity restrictions. Then, the matched RESTful interfaces from the vector library are placed into the parameter block of the function-calling module, and the large language model selects the most suitable interface resource.

[0092] The same approach is used when executing the next segment, but the execution result of the previous segment is added to the dialog context. This process is repeated until all segments are executed, and then the execution result of the last segment is returned. This method allows for the assembly of services in natural language, and can be extended to include other types of services in the future.

[0093] Based on the same inventive concept, embodiments of the present invention also propose a readable storage medium on which a computer program is stored, and when the computer program is executed, it can implement the software development method based on natural language arrangement as described above.

[0094] A readable storage medium can be a tangible device capable of holding and storing instructions for use by an instruction execution device, such as, but not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer programs described herein can be downloaded from the readable storage medium to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. Networks can include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. Each computing / processing device's network adapter card or network interface receives and forwards a computer program from the network for storage on a readable storage medium within the respective computing / processing device. The computer program used to perform the operations of this invention can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as "C" or similar languages. The computer program can execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuits, such as programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), are personalized by utilizing state information from a computer program. These electronic circuits can execute computer-readable program instructions, thereby realizing various aspects of the present invention.

[0095] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by a computer program. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. These computer programs can also be stored in a readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the readable storage medium storing the computer program comprises an article of manufacture including instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams.

[0096] A computer program may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the computer program executing on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0097] In summary, the embodiments of the present invention provide a software development method, system, and readable storage medium based on natural language orchestration. This allows users to break down business requirements into detailed steps using natural language without needing to understand the technical principles, which are then executed by a large language model. This method achieves natural language orchestration, lowers the barrier to entry, allows users to complete software development independently, reduces communication time, and reduces the possibility of requirement deviations.

[0098] Furthermore, it should be understood that although the present invention has been disclosed above with reference to preferred embodiments, these embodiments are not intended to limit the present invention. For any person skilled in the art, many possible variations and modifications can be made to the technical solutions of the present invention based on the disclosed technical content, or equivalent embodiments can be modified accordingly, without departing from the scope of the present invention. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention, without departing from the content of the present invention, shall still fall within the scope of protection of the present invention.

Claims

1. A software development method based on natural language orchestration, characterized in that, Includes the following steps: S1. Deploy a programmable system based on natural language orchestration, register basic resources in the programmable system, describe the registered resources and system resources in natural language, store them in a relational database, and synchronize them to a vector database. S2. Create and orchestrate business applications in the programmable system, then describe the orchestration content in natural language, publish it as an intelligent agent, store it in the relational database, and synchronize it to the vector database. S3. Use a large language model to perform vector retrieval on the user's question content and select the most suitable agent; S4. Call the most suitable agent to execute the corresponding arrangement content in paragraph order and output the final result; Specifically, step S3 includes: The large language model is used to perform vector retrieval on the question content to identify multiple agents; The multiple agents are placed into the parameter block of the Function calling module of the large language model, and the large language model selects the most suitable agent.

2. The software development method based on natural language arrangement according to claim 1, characterized in that, In S1, the registration of the basic resources includes the registration of interface resources and the publication of database resources.

3. The software development method based on natural language arrangement according to claim 1, characterized in that, In step S1, the registered resources and system resources after natural language description are first converted into SQL statements. Then, the SQL statements are published as a standard RESTful interface and stored in the relational database and synchronized to the vector database.

4. The software development method based on natural language arrangement according to claim 1, characterized in that, The steps in S4 specifically include: S41. Call the most suitable intelligent agent to perform vector retrieval on the content of each paragraph in the order of the paragraphs, and find multiple interface resources; S42. Place the multiple interface resources into the parameter block of the Function calling module, and the large language model selects the most suitable interface resource. S43. Extract parameters from the question content and the returned content of the preceding paragraph, and use the extracted parameters to call the most suitable interface resource; S44. Repeat S41-S43 until all paragraphs have been executed; S45. Output the final result.

5. A programmable system based on natural language arrangement, characterized in that, include: The data management module is used for registering basic resources; The resource management module is used to describe registered resources and system resources in natural language, store them in a relational database, and synchronize them to a vector database. Human-computer interaction interface, used to interact with users and create business applications; The orchestration module is used to orchestrate the business application, and then describe the orchestration content in natural language, publish it as an intelligent agent, store it in the relational database, and synchronize it to the vector database. The large language model is used to perform vector retrieval on the user's query content, select the most suitable agent, and call the most suitable agent to execute the corresponding arrangement content in paragraph order, and output the final result. The large language model includes a function calling module, which is specifically used for: Vector retrieval is performed on the query content to find multiple agents, and these agents are placed into the parameter block of the Function calling module to select the most suitable agent for invocation and execution.

6. The programmable system based on natural language arrangement according to claim 5, characterized in that, The programmable system also includes a text2sql module, which is used to convert the registered resources and system resources after natural language description into SQL statements, publish the SQL statements as a standard RESTful interface, store them in the relational database, and synchronize them to the vector database.

7. The programmable system based on natural language arrangement according to claim 5, characterized in that, The large language model is also specifically used for: The most suitable agent is invoked to perform vector retrieval on the content of each paragraph in the order of the paragraphs, and multiple interface resources are found. The multiple interface resources are placed into the parameter block of the Function calling module to select the most suitable interface resource; The parameters are extracted from the question content and the returned content of the preceding paragraphs, and the extracted parameters are used to call the most suitable interface resource until all paragraphs are executed. Then the final result is output to the human-computer interaction interface.

8. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it can implement the software development method based on natural language arrangement according to any one of claims 1-4.

Images