Program building system based on user needs

Abstract

The program building system based on user needs proposed in this application relates to the field of system architecture and design technology. It reads the open source file library of the low-code platform; based on the program building type, the open source file library is clustered and segmented to determine multiple open source sub-libraries; the program building type is traversed to determine the underlying logic of the program, a building guidance model is constructed, and a direct calling channel between the building guidance model and multiple open source sub-libraries is established; user needs are read, the building guidance model is assisted, and the open source code call adjustment and combination based on multiple open source sub-libraries are combined with the direct calling channel to generate a target program. This solves the technical problems in the existing technology that the software is difficult to adapt to changes in demand during the development and design process, the development is difficult, the efficiency is low, and it is difficult to maintain, resulting in waste of resources. It achieves the technical effect of quickly and efficiently utilizing open source resources for program building on the low-code platform, reducing development difficulty and improving development efficiency.

Description

Technical Field

[0001] The present invention relates to the field of system architecture and design technology, and in particular to a program building system based on user needs. Background Art

[0002] In an environment of globalized markets and personalized user demands, quickly and accurately understanding and meeting user needs is crucial for business survival and growth. The microservices architecture allows software systems to be divided into a series of small services, each running in its own independent process and using lightweight communication mechanisms between services. This architecture makes the system more flexible and scalable, and better able to adapt to changing user needs. However, during the software design process, the analysis of user needs is often incomplete and incomplete, resulting in a poor match between software development and design, making it difficult to adapt to changing needs, increasing development complexity, inefficiency, and difficulty in maintenance, leading to a waste of manpower, computing power, and time resources.

[0003] In summary, existing technologies often have technical problems such as low matching degree between software and actual needs during the development and design process, difficulty in adapting to changes in needs, high development difficulty, low efficiency and difficulty in maintenance, resulting in waste of manpower, computing power and time resources. Summary of the Invention

[0004] This application provides a program building system based on user needs, which is used to solve the technical problems in the existing technology that the software has a low match with actual needs during the development and design process, is difficult to adapt to changes in needs, is difficult to develop, has low efficiency and is difficult to maintain, resulting in a waste of manpower, computing power and time resources.

[0005] This application provides a program building system based on user needs, which includes:

[0006] A file reading module, used to read the open source file library of the application platform, where the application platform is a low-code platform;

[0007] A clustering and segmentation module, configured to perform open source clustering and segmentation on the open source file library based on the program construction type, and determine multiple open source sub-libraries;

[0008] A model building module, configured to traverse the program building type, determine the underlying program logic, and construct a building guidance model based on the underlying program logic, wherein the building guidance model includes a plurality of guidance branches, and the plurality of guidance branches correspond one to one with the program building type;

[0009] A channel establishment module, used to establish a direct calling channel between the construction guidance model and the multiple open source sub-libraries;

[0010] The program generation module is used to read user needs, assist in building the guidance model, combine the direct call channel, adjust and combine the open source code calls based on the multiple open source sub-libraries, and generate a target program.

[0011] The program building system based on user needs proposed in this application reads the open source file library of the application platform, which is a low-code platform; based on the program building type, the open source file library is clustered and segmented to determine multiple open source sub-libraries; the program building type is traversed to determine the underlying logic of the program, and a building guidance model is constructed based on the underlying logic of the program, the building guidance model includes multiple guidance branches, and the multiple guidance branches correspond one-to-one to the program building type; a direct calling channel is established between the building guidance model and the multiple open source sub-libraries; user needs are read to assist the building guidance model, and in combination with the direct calling channel, open source code calls based on the multiple open source sub-libraries are adjusted and combined to generate a target program, which solves the technical problems in the existing technology that the software has a low matching degree with actual needs during the development and design process, is difficult to adapt to changes in needs, has high development difficulty, low efficiency and is difficult to maintain, resulting in a waste of manpower, computing power and time resources, and achieves the technical effect of quickly and efficiently utilizing open source resources for program building on a low-code platform, reducing development difficulty and improving development efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of the structure of a program building system based on user needs provided in an embodiment of the present application.

[0013] Description of the accompanying drawings: file reading module 11, clustering and segmentation module 12, model building module 13, channel establishment module 14, program generation module 15. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0015] The present application embodiment provides a program building system based on user needs, such as Figure 1 As shown, the system includes:

[0016] A file reading module 11 is used to read the open source file library of the application platform, where the application platform is a low-code platform;

[0017] Specifically, the file reading module is used to read the open source file library of the low-code application platform to lay the foundation for the generation of the target program. Among them, the low-code platform refers to a development environment that uses a small amount of handwritten code and a large number of graphical interface operations to generate applications. The open source file library refers to a file library whose source code is freely available to the public for acquisition, use, modification and redistribution. These file libraries usually follow specific open source licenses, such as MIT, Apache2.0, BSD, etc. By reading the specific location and access method of the open source file library of the low-code platform, the source code of the open source file library can be obtained from the designated code hosting platform according to the demand information provided by the platform.

[0018] By building programs based on a low-code platform, the difficulty of learning programming languages is greatly reduced, development efficiency is improved, and labor costs are reduced.

[0019] A clustering and segmentation module 12 is configured to perform open source clustering and segmentation on the open source file library based on the program construction type, and determine a plurality of open source sub-libraries;

[0020] Optionally, collect all open source repositories related to the program build type, such as by retrieving and downloading relevant code repositories from multiple code hosting platforms. Use code parsing tools to parse the source code in the open source repositories, extracting key elements such as functions, classes, and modules. Based on the code parsing results, identify the function and purpose of each key element. Also, extract metadata related to the code repositories, such as version, license, and dependencies. Next, select an appropriate clustering algorithm to cluster the open source repositories. The choice of clustering algorithm depends on the specific application scenario and data characteristics. Common clustering algorithms include K-means, hierarchical clustering, and DBSCAN. From the key elements and metadata extracted during the preprocessing and analysis phases, extract clustering features. These features include code repositories' size, complexity, dependencies, and function labels. Select the most representative feature subset based on the clustering algorithm's needs. Next, use the features from the extracted feature subset as input and apply the selected clustering algorithm to cluster the open source repositories. Based on the clustering results, analyze the common characteristics and differences between the open source repositories in each cluster to determine whether they can be considered distinct open source sub-repositories. Based on the results of the cluster analysis, identify multiple open source sub-repositories. Each sub-repository should contain open source document libraries with similar functions and features. Assign each open source sub-repository a unique identifier (such as sub-repository name, version number, etc.) and generate corresponding documentation and metadata for subsequent use and management.

[0021] Through clustering and segmentation, developers can more easily find the required code snippets and improve development efficiency. In addition, the formation of multiple open source sub-libraries helps to optimize the management of the code base and facilitate subsequent maintenance and updates.

[0022] A model building module 13 is used to traverse the program building type, determine the program underlying logic, and construct a building guidance model based on the program underlying logic, wherein the building guidance model includes multiple guidance branches, and the multiple guidance branches correspond to the program building type one by one;

[0023] For example, clarify and define the types of program builds to be handled, including different frameworks, libraries, design patterns, architectural styles, etc. For each build type, collect a certain number of sample codes or projects for subsequent analysis. Next, use code analysis tools (such as built-in functions of IDEs, static code analysis tools, etc.) to deeply understand the underlying logic of each sample code, decompose the code according to its function, identify key components, modules, functions and the interactions between them, and abstract general, underlying logical structures from the specific code. These structures are generally independent of specific implementation details.

[0024] Furthermore, based on the underlying logic, a guidance model with multiple guidance branches is designed. Each branch should correspond to a specific program build type. Specific content is defined for each guidance branch, including the characteristics of the build type, applicable scenarios, required resources, and build steps. This ensures that each guidance branch is closely linked to the underlying logic of the corresponding program build type, providing accurate guidance during the guidance process.

[0025] The build guidance model provides developers with clear program building instructions, reducing development difficulty. Furthermore, multiple guidance branches correspond to program building types, enabling the system to flexibly respond to diverse development needs. Furthermore, the build guidance model, built based on the underlying program logic, ensures the rationality and stability of the program structure.

[0026] A channel establishment module 14 is used to establish a direct calling channel between the building guidance model and the multiple open source sub-libraries;

[0027] Further, a direct call channel between the construction guide model and the multiple open source sub-libraries is established so that users or developers can directly quote or call the relevant open source sub-libraries when using the construction guide model to build a program. Specifically, the interaction mode between the construction guide model and the open source sub-libraries is clearly defined, and a unified interface specification is defined so that different open source sub-libraries can be uniformly called in the construction guide model. Existing dependency management tools (such as npm, yarn, Maven, Gradle, etc.) are used to manage the references and dependencies of the open source sub-libraries, and these tools are integrated in the construction guide model so that the download, installation and version management of the sub-libraries can be automatically handled during the construction process. In the implementation of the construction guide model, the corresponding open source sub-libraries are dynamically loaded and called according to the user's choice and construction type. At the same time, the call channel between the construction guide model and the open source sub-libraries is fully tested and verified to ensure the correctness and stability of the call process, and different construction types and sub-library combinations are used for testing to ensure that the system can correctly handle various situations and errors.

[0028] The established direct call channel allows the bootstrap model to directly access and call code in the open-source sub-repository at runtime, significantly improving development efficiency. It also reduces coupling between modules, making the system easier to maintain and expand. Furthermore, when code in the open-source sub-repository is updated, the channel reflects these changes in real time, ensuring the real-time performance and accuracy of the system.

[0029] The program generation module 15 is used to read user requirements, assist in building the guidance model, and combine the direct call channel to adjust and combine the open source code calls based on the multiple open source sub-libraries to generate a target program.

[0030] Specifically, clear and accurate reading and understanding of user requirements can be achieved through user interface (UI) forms, dialog boxes, questionnaires, or other input methods. Furthermore, the user's input requirements are converted into a form understandable by the program, such as data structures, algorithm requirements, and function lists. Based on the user's needs, the build guidance model can provide a series of suggestions or options to guide the user in selecting the appropriate program build type and open source sub-library. Using machine learning or a rule engine, the system can intelligently recommend the build type and sub-library combination that best meets the user's needs based on historical data and user behavior. Next, based on the user's selection and the build guidance model's recommendations, the system determines the open source sub-libraries to call and their dependencies. Based on these determined call relationships, a direct call channel is configured between the build guidance model and the open source sub-libraries to ensure that these sub-libraries are correctly called when the target program is generated. Furthermore, based on the user's needs and the specific requirements of the program build, necessary adjustments are made to the code called from the open source sub-libraries, such as modifying parameters, adding functionality, or fixing errors. These adjusted code fragments are then combined according to a predetermined logic and sequence to form a complete program flow. Compile the combined code to ensure its correctness and operability. At the same time, perform detailed tests on the compiled program, including unit testing, integration testing, functional testing, etc., to ensure that the program meets user requirements. The program that passes the test is packaged as an executable file, library file or other form of output and provided to users.

[0031] During the generation process, the module optimizes and combines the code to ensure the quality and stability of the target program. At the same time, the generated target program not only meets functional requirements, but also focuses on user experience, improving user satisfaction.

[0032] Furthermore, to obtain the program building type, the cluster segmentation module 12 is further configured to:

[0033] Based on a preset period, reading the program construction record of the application platform;

[0034] Traversing the program construction records, performing a clustering based on demand to determine a first clustering result, and performing a second clustering based on open source data to determine a second clustering result;

[0035] fusing the first clustering result and the second clustering result to determine a program clustering result;

[0036] Traverse the program clustering results, identify the types within the cluster, and determine the program construction type.

[0037] Optionally, within a preset period (such as daily, weekly, or monthly), program build records are read from the application platform's database. These records typically contain detailed information about the program build, such as build time, user information, open source libraries used, and build requirements. All program build records are traversed to extract features related to build requirements (such as functional requirements, performance requirements, and interface requirements). These features are clustered using a clustering algorithm (such as K-means or hierarchical clustering). Program build records with similar requirements are grouped together to form a first clustering result. The program build records are traversed again to extract features related to open source data (such as open source libraries used, library versions, and dependencies between libraries). These features are clustered using a clustering algorithm. Program build records that use similar open source data or have similar dependencies are grouped together to form a second clustering result. The first clustering result and the second clustering result are fused. The fusion process needs to consider factors such as the degree of overlap between the two clustering results, the distance between cluster centers, and the similarity within the cluster. The specific fusion process can be performed using weighted fusion to determine the final program clustering result. Next, we traverse the fused program clustering results and identify the type within each cluster. We then determine the program type for each cluster based on the characteristics of the program build records within the cluster (such as requirements and open source data). The program build type can be a specific classification label (such as a web application, data analysis tool, or game) or a more granular description (such as a web application based on a specific framework or a data analysis tool using a specific algorithm).

[0038] Through the above steps, we can achieve a comprehensive understanding and optimization support of the program building process, providing developers with a more efficient, convenient and personalized development experience.

[0039] Furthermore, the open source file library is extensible and updatable, and the multiple open source sub-libraries are extensible and updatable.

[0040] Specifically, open source libraries and the multiple open source sub-libraries they contain are subject to extensible updates, meaning they are subject to updates over time to add new features or fix issues. When working with these extensible libraries, ensure that your systems and processes can flexibly adapt to these changes. Optionally, track the latest version and changelog for each open source library and sub-library, and explicitly specify the library and sub-library versions used within your project to ensure consistency. Integrate dependency management tools into your project's build process to automatically download and install the required library and sub-library versions during the build. Also, set up regular tasks or scripts to automatically check for new library or sub-library releases and assess whether updates are necessary. Before integrating a new version into your project, conduct thorough testing to ensure compatibility. If a new version introduces issues or doesn't meet expectations, ensure a rollback mechanism to the previous version. When open source libraries or sub-libraries are updated, analyze whether these changes impact the recommendations and configurations of the build boot model. Based on the analysis, update the build boot model to reflect the latest library and sub-library versions and possible best practices. Additionally, consider user notifications, documentation and maintenance, security considerations, and backup and recovery.

[0041] Furthermore, the multiple open source sub-libraries include extended updates based on sub-library divisions and extended updates based on data within the library.

[0042] For example, expansion and update based on sub-library division can achieve a more refined modular design by splitting a large open source library into multiple independent sub-libraries. Each sub-library is responsible for a specific function or business area, making the system clearer, easier to understand and maintain. When business needs change or new technologies emerge, developers can independently update or expand specific sub-libraries without making large-scale modifications to the entire system. This flexibility enables the system to adapt to changes more quickly. When expansion and update are required, optimization is completed by adding new sub-libraries or updating existing sub-libraries. For example, as business needs grow, new sub-libraries may need to be added to support new functions or business areas. These new sub-libraries can be implemented based on the existing technology stack or by introducing new technologies; or when the functions in the sub-library need to be improved, repaired or expanded, developers can update specific sub-libraries, including adding new functions, fixing known problems or optimizing performance.

[0043] Open-source sub-databases typically contain a large amount of data and business logic. This data is fundamental to the sub-database's operation and functionality. Therefore, maintaining and updating the data within the database is crucial. As business needs change and technology advances, database data may require adjustments and optimization to ensure the sub-database continues to provide efficient, accurate, and reliable services. Database data expansion and updates can include data model updates, data processing optimizations, and data migration and synchronization. For example, when business needs change, the sub-database's data model may need to be updated to accommodate new data structures or business logic, such as adding new fields, modifying field types, or adjusting data relationships. Alternatively, to improve the sub-database's performance and responsiveness, developers may need to optimize data processing algorithms, including using more efficient data structures, improving algorithmic logic, or incorporating parallel processing techniques. Furthermore, when a sub-database needs to be migrated to a new storage system or exchange data with other systems, data migration and synchronization may be necessary, such as migrating data from an old system to a new one and ensuring data consistency and accuracy across multiple systems.

[0044] Two key aspects of open source sub-library maintenance and updates are expansion and updates based on sub-library partitioning and internal data. The former focuses on the overall system architecture and organization, improving system flexibility and scalability through modular design; the latter focuses on the data and business logic within the sub-library, ensuring that the sub-library can continue to provide efficient, accurate, and reliable services through data-driven and continuous iteration. In practical applications, developers can choose the appropriate expansion and update method based on their specific needs and circumstances to meet the continuous development and changes of the system.

[0045] Furthermore, based on the underlying logic of the program, the model building module 13 is further used to build a guidance model:

[0046] Traversing the program building types, extracting the first type, and determining the first program logic based on the response results and the program structure;

[0047] Building a first boot branch based on the first program logic;

[0048] Identifying the program construction record and determining a first training sample based on the first type;

[0049] Based on the first training sample, supervised training is performed on the first guiding branch to meet a preset convergence degree, so as to obtain a constructed first guiding branch.

[0050] Furthermore, the system traverses all possible program construction types, distinguishing them based on different programming languages, frameworks, design patterns, or business logic. During this traversal, a specific first type is extracted, selected based on a specific need, scenario, or condition. For each of these extracted first types, the system analyzes the response results and program structure of the program within that type. By analyzing these results and structure, the system determines the first program logic, which involves a series of steps, conditional judgments, and function calls to achieve a specific function or business goal. Using this determined first program logic, the system then begins building a first guidance branch. This guidance branch may be a user interface component, part of a decision tree, or a specific module of an AI model. The primary function of the guidance branch is to guide or direct users, developers, or other systems to perform operations or make decisions according to the first program logic. After building the first guidance branch, the system identifies program construction records related to the first type. These records include historical data, user feedback, test cases, and so on. Data closely related to the first type is then filtered from these records to serve as the first training sample. Next, supervised training of the first guidance branch is performed using the first training sample. During the training process, the system evaluates the performance of the model and continuously adjusts the model parameters to improve its accuracy. This process continues until the model reaches a preset convergence. When the first guidance branch meets the preset convergence, it is considered to be constructed. At this point, the guidance branch has the ability to accurately guide and make decisions according to the first program logic. The constructed first guidance branch can be integrated into a larger system or application to guide users, developers, or other systems to perform operations or make decisions according to specific program logic.

[0051] Through in-depth analysis of specific program building types, construction of guide branches, and supervised training based on real training samples, we ultimately build accurate, stable, and practical guide branches, providing developers with efficient and standardized program building support.

[0052] Furthermore, the model building module 13 is also used to:

[0053] Based on the program construction type and the program construction record, completing the training of the Nth guide branch;

[0054] The first guiding branch up to the Nth guiding branch are configured in parallel to generate the building guiding model, wherein each guiding branch has an independent calculation domain.

[0055] Optionally, the training of the Nth guiding branch is completed based on the same logic as the training of the first guiding branch. After all guiding branches are trained, they are configured in parallel. That is, they will run simultaneously, but each will handle a different program setup type. This parallel configuration ensures that the system can handle multiple different program setup requirements simultaneously, improving overall processing efficiency and response speed. Furthermore, the parallel configured guiding branches are integrated into a complete setup guidance model, which will automatically select the corresponding guiding branch for processing based on the input program setup type. At the same time, the setup guidance model also needs to ensure that the guiding branches work together to ensure the consistency and accuracy of the entire model. Furthermore, during the generation of the setup guidance model, it is necessary to ensure that each guiding branch has its own independent computational domain. That is, each branch will not interfere with each other during operation and will process its own data and logic. Independent computational domains ensure the stability and reliability of each guiding branch, preventing the failure or error of a branch from affecting the operation of the entire model. Furthermore, after the generation of the setup guidance model is completed, it is necessary to conduct thorough testing to ensure its accuracy and reliability. Based on the test results, the model should be optimized and adjusted to improve its performance and accuracy.

[0056] Furthermore, each guiding branch has an independent calculation domain, and the model building module 13 is further used to:

[0057] Evenly distribute the branch computing power of the built guidance model and determine the initialization distribution result;

[0058] Based on the program construction records and the construction frequency of each level of construction type, an adjustment ratio is determined;

[0059] Based on the adjustment ratio, the computing power domain of the initialization allocation result is adjusted.

[0060] For example, the computing power required for normal operation of each boot branch is assessed. This can be determined through simulations, historical data, or expert evaluation. After understanding the computing power requirements of each branch, the total computing power resources are allocated to each branch in a balanced manner to ensure that each branch receives sufficient computing power to support its normal operation. The computing power allocation results for each branch are recorded and used as a benchmark for subsequent adjustments. Furthermore, a detailed analysis of program build records is performed, particularly the build frequency of each build type. This can be used to understand which build types are more popular and which are less popular. Furthermore, based on the build frequency, the ratio of each build type relative to other types is calculated. This ratio is used to adjust the computing power allocation results. Next, based on the calculated adjustment ratio, the computing power allocation of each branch is determined. Generally speaking, more popular build types (i.e., those with higher build frequency) should receive more computing power support. The computing power allocation results are adjusted according to the adjustment strategy. Specifically, this can be achieved by increasing or decreasing the computing power quota of specific branches. After the adjustment, a series of verification tests are conducted to ensure that the new computing power allocation meets the needs of various build types and does not cause performance degradation in certain branches due to insufficient computing power.

[0061] Through the above steps, we can achieve balanced distribution of computing power among the branches of the boot model and adjust it based on actual needs. This improves the overall performance and stability of the boot model and ensures that key or commonly used branches have sufficient computing power resources to meet actual needs. Furthermore, this adjustment method based on actual usage data also makes the entire boot model more consistent with the needs of actual development scenarios, improving its practicality and effectiveness in real-world applications.

[0062] Furthermore, the system further includes a learning update module, which is configured to:

[0063] Recording user behavior, wherein the user behavior refers to operational behavior that is different from program construction guided by the model;

[0064] Conduct a standardization assessment on the user behavior, and if it meets the construction standard, it is considered as a valid learning behavior;

[0065] Based on a preset period, the effective learning behavior is identified and the building guidance model is updated through learning.

[0066] Specifically, define user behavior, that is, clarify which operational behaviors are different from those of model-guided program construction. These behaviors may be additional operations or modifications performed by users independently and independent of model guidance. Furthermore, implement a mechanism to track and record these user behaviors. Specifically, this can be accomplished through logging, event tracking, or a dedicated user behavior analysis system. Ensure that the collected user behavior data is comprehensive, accurate, and analyzable. Furthermore, establish user behavior standards, that is, clarify what behaviors meet the construction standards. This may need to be based on business logic, programming specifications, best practices, or expert opinions. Next, analyze the user behavior data to determine whether each behavior meets the preset standards. Behaviors that meet the standards are marked as valid learning behaviors, while behaviors that do not meet the standards are not included in the subsequent learning and update process.

[0067] Furthermore, a preset cycle, such as daily, weekly, or monthly, is established to regularly identify and analyze effective learning behaviors. At the end of each cycle, effective learning behaviors are screened from the recorded user behavior data and analyzed in depth to understand their patterns, trends, and potential impacts. Based on the analysis results of effective learning behaviors, an update strategy for the onboarding model is formulated, including adjusting model parameters, optimizing algorithms, adding new features, or improving the user interface. The onboarding model is trained using effective learning behaviors as training data. After training, the updated model is verified and tested to ensure that its performance and accuracy meet the expected standards. The verified updated model is deployed in a live environment for user use and testing.

[0068] In summary, recording user behavior, assessing its standardization, and updating the guidance model based on effective learning not only improves user experience and satisfaction, but also optimizes the accuracy and applicability of the guidance model, enabling self-iteration and optimization of the model, promoting communication and knowledge sharing, and enhancing the platform's adaptability and scalability. These effects collectively drive the platform toward greater intelligence and personalization.

[0069] Through the technical solutions of the above embodiments, the program building system based on user needs provided by this application solves the technical problems in the prior art that the software has a low degree of matching with actual needs during the development and design process, is difficult to adapt to changes in needs, has high development difficulty, low efficiency and is difficult to maintain, resulting in a waste of manpower, computing power and time resources. It achieves the technical effect of quickly and efficiently utilizing open source resources to build programs on a low-code platform, reducing development difficulty and improving development efficiency.

[0070] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A program building system based on user needs, characterized by: The system comprises: A file reading module, used to read the open source file library of the application platform, where the application platform is a low-code platform; A clustering and segmentation module, configured to perform open source clustering and segmentation on the open source file library based on the program construction type, and determine multiple open source sub-libraries; A model building module, configured to traverse the program building type, determine the underlying program logic, and construct a building guidance model based on the underlying program logic, wherein the building guidance model includes a plurality of guidance branches, and the plurality of guidance branches correspond one to one with the program building type; A channel establishment module, used to establish a direct calling channel between the construction guidance model and the multiple open source sub-libraries; A program generation module is used to read user requirements, assist in building the guidance model, and combine the open source code calls based on the multiple open source sub-libraries in combination with the direct call channel to generate a target program; The cluster segmentation module is further used for: Based on a preset period, reading the program construction record of the application platform; Traversing the program construction records, extracting features related to construction requirements, clustering these features using a clustering algorithm, and grouping program construction records with similar requirements into one category to form a first clustering result; traversing the program construction records again, extracting features related to open source data, clustering these features using a clustering algorithm, and grouping program construction records that use similar open source data or have similar dependencies into one category to form a second clustering result; fusing the first clustering result with the second clustering result to determine a program clustering result; Traverse the program clustering results, identify the types within the cluster, and determine the program construction type.

2. The program building system based on user needs according to claim 1, characterized in that: The open source file library is extensible and updatable, and the multiple open source sub-libraries are extensible and updatable.

3. The program building system based on user needs according to claim 2, characterized in that: The multiple open source sub-libraries include extended updates based on sub-library divisions and extended updates based on data within the library.

4. The program building system based on user needs according to claim 1, characterized in that: The model building module is also used to: Traversing the program building types, extracting the first type, and determining the first program logic based on the response results and the program structure; Building a first boot branch based on the first program logic; Identifying the program construction record and determining a first training sample based on the first type; Based on the first training sample, supervised training is performed on the first guiding branch to meet a preset convergence degree, so as to obtain a constructed first guiding branch.

5. The program building system based on user needs according to claim 4, characterized in that: The model building module is also used to: Based on the program construction type and the program construction record, completing the training of the Nth guide branch; The first guiding branch up to the Nth guiding branch are configured in parallel to generate the building guiding model, wherein each guiding branch has an independent calculation domain.

6. The program building system based on user needs according to claim 5, characterized in that: The model building module is also used to: Evenly distribute the branch computing power of the built guidance model and determine the initialization distribution result; Based on the program construction records and the construction frequency of each level of construction type, an adjustment ratio is determined; Based on the adjustment ratio, the computing power domain of the initialization allocation result is adjusted.

7. The program building system based on user needs according to claim 1, characterized in that: The system further includes a learning update module, which is configured to: Recording user behavior, wherein the user behavior refers to operational behavior that is different from program construction guided by the model; The user behavior is judged to be of standard nature, and if it meets the construction standard, it is regarded as a valid learning behavior; based on a preset period, the valid learning behavior is identified, and the construction guidance model is updated through learning.

Images