Program building system based on user requirements

Abstract

The invention provides a program building system based on user requirements, and relates to the technical field of system architecture and design, and the program building system comprises the following steps: reading an open source file library of a low-code platform; performing open source clustering and segmentation on the open source file library based on the program building type, and determining a plurality of open source sub-libraries; traversing program building types, determining program underlying logic, building a building guide model, and building direct calling channels of the building guide model and the open source sub-libraries; the method comprises the steps of reading user requirements, assisting in establishing a guide model, performing open source code calling adjustment and combination based on a plurality of open source sub-libraries in combination with a direct calling channel, and generating a target program, and solves the problems that in the prior art, software is difficult to adapt to requirement changes in the development and design process, the development difficulty is high, the efficiency is low, and maintenance is not easy; and the technical effects of quickly and efficiently utilizing the open source resources to carry out program establishment on the low-code platform, reducing the development difficulty and improving the development efficiency are achieved.

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 market globalization and personalized user needs, quickly and accurately understanding and meeting user needs is the key to the survival and development of enterprises. The microservice architecture enables the software system to be divided into a series of small services, each of which runs in its own independent process, and a lightweight communication mechanism is used between services. This architecture makes the system more flexible and scalable, and can better meet the changes in user needs. However, the analysis of user needs in the software design process is often not comprehensive and thorough, which makes the software poorly matched with actual needs during the development and design process, and difficult to adapt to changes in needs. It is difficult to develop, inefficient and difficult to maintain, resulting in a waste of manpower, computing power and time resources.

[0003] In summary, the existing technology often has technical problems such as the software has a low match with actual needs during the development and design process, and is difficult to adapt to changes in needs. It is difficult to develop, inefficient and difficult to maintain, resulting in a waste of manpower, computing power and time resources. Summary of the invention

[0004] The present application provides a program building system based on user needs, which is used to solve the technical problems in the prior art that the software has a low match 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.

[0005] This application provides a program building system based on user needs, the system 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, used 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, used to traverse the program building type, determine the underlying logic of the program, and construct a building guidance model based on the underlying logic of the program, wherein the building guidance model includes a plurality of guidance branches, and the plurality of guidance branches correspond to the program building type one by one;

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

[0010] The program generation module 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.

[0011] The program building system based on user needs proposed in this application is proposed. By reading the open source file library of the application platform, the application platform 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 to the program building type one by one; a direct calling channel between the building guidance model and the multiple open source sub-libraries is established; 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 prior art 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 not easy to maintain, resulting in a waste of manpower, computing power and time resources, and achieves the technical effect of quickly and efficiently using open source resources to build programs 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] Explanation 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 solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

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

[0016] A file reading module 11, used to read an open source file library of an 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 used 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;

[0020] Optionally, collect all open source file libraries related to the program building type, such as retrieving and downloading relevant code libraries from multiple code hosting platforms. Use code parsing tools to parse the source code in the open source file library, extract key elements such as functions, classes, modules, and identify the functions and uses of each key element based on the results of code parsing, and extract metadata related to the code library, such as version, license, dependency, etc. Next, select a suitable clustering algorithm to cluster the open source file library. The selection of clustering algorithm depends on the specific application scenario and data characteristics. Common clustering algorithms include K-means, hierarchical clustering, DBSCAN, etc. Extract features for clustering from the key elements and metadata extracted in the preprocessing and analysis stages. These features include the size, complexity, dependency, function label, etc. of the code library, and select the most representative feature subset according to the needs of the clustering algorithm. Next, use the features in the extracted feature subset as input, apply the selected clustering algorithm to cluster the open source file library, and analyze the common characteristics and differences of the open source file libraries in each cluster based on the clustering results to determine whether they can be regarded as different open source sub-libraries. According to the results of cluster analysis, multiple open source sub-libraries are determined. Each sub-library should contain open source file libraries with similar functions and features. Each open source sub-library is assigned a unique identifier (such as sub-library name, version number, etc.) and corresponding documents and metadata are generated 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, used to traverse the program building type, determine the underlying logic of the program, and construct a building guidance model based on the underlying logic of the program, 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 construction to be handled, including different frameworks, libraries, design patterns, architectural styles, etc. For each construction type, collect a certain number of sample codes or projects for subsequent analysis. Next, use code analysis tools (such as built-in functions of IDE, 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 their interactions, and abstract the general, underlying logical structure from the specific code, which usually does not depend on the specific implementation details.

[0024] Furthermore, a guidance model with multiple guidance branches is designed based on the underlying logic, and each branch should correspond to a specific program building type. Specific content is defined for each guidance branch, including the characteristics of the building type, applicable scenarios, required resources, building steps, etc., to ensure that each guidance branch is closely related to the underlying logic of the corresponding program building type, so as to provide accurate guidance during the guidance process.

[0025] The building guidance model provides developers with clear program building guidance, reducing the difficulty of development. In addition, multiple guidance branches correspond to program building types one by one, allowing the system to flexibly respond to different development needs. At the same time, the building guidance model built based on the underlying logic of the program 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 guide model and the plurality of 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 reference or call related open source sub-libraries when using the construction guide model to build programs. Specifically, the interaction mode between the construction guide model and the open source sub-libraries is clarified, and a unified interface specification is defined so that different open source sub-libraries can be uniformly called in the construction guide model. Use existing dependency management tools (such as npm, yarn, Maven, Gradle, etc.) to manage the references and dependencies of the open source sub-libraries, and integrate these tools in the construction guide model to automatically handle the download, installation and version management of the sub-libraries during the construction process. In the implementation of the construction guide model, according to the user's choice and construction type, the corresponding open source sub-libraries are dynamically loaded and called. At the same time, the call channel between the construction guide model and the open source sub-library 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 construction guide model to directly access and call the code in the open source sub-library at runtime, which greatly improves the development efficiency. At the same time, the direct call channel reduces the coupling between modules, making the system easier to maintain and expand. In addition, when the code in the open source sub-library is updated, the channel can reflect these changes in real time, ensuring the real-time 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, the user's needs can be clearly and accurately read and understood through the form, dialog box, questionnaire or other input methods of the user interface (UI). Further, the user input needs are converted into a form that the program can understand, such as data structure, algorithm requirements, function list, etc. Then, according to the user's needs, the building guidance model can provide a series of suggestions or options to guide the user to choose the appropriate program building type and open source sub-library. The system can intelligently recommend the building type and sub-library combination that best meets the user's needs based on historical data and user behavior through machine learning or rule engines. Next, based on the user's choice and the recommendation of the building guidance model, the open source sub-libraries that need to be called and the dependencies between them are determined, and according to the determined calling relationship, the direct calling channel between the building guidance model and the open source sub-library is configured to ensure that these sub-libraries can be correctly called when the target program is generated. Then, according to the user's needs and the specific requirements of the program building, the necessary adjustments are made to the code called from the open source sub-library, such as modifying parameters, adding functions, fixing errors, etc., and the adjusted code fragments are combined according to the predetermined logic and order to form a complete program flow. Compile the combined code to ensure the correctness and operability of the code. 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 needs. Package the tested program into executable files, library files or other forms of output for user use.

[0031] During the generation process, the module will optimize the code combination to ensure the quality and stability of the target program. At the same time, the generated target program not only meets the 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 used to:

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

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

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

[0036] The program clustering results are traversed to identify the types within the cluster and determine the program building type.

[0037] Optionally, within a preset period (such as daily, weekly, monthly, etc.), read the program construction records from the database of the application platform. These records usually contain detailed information about the program construction, such as construction time, user information, open source libraries used, construction requirements, etc. Traverse all program construction records, extract features related to construction requirements (such as functional requirements, performance requirements, interface requirements, etc.), use clustering algorithms (such as K-means, hierarchical clustering, etc.) to cluster these features, and classify program construction records with similar requirements into one category to form a first clustering result. Traverse the program construction records again, extract features related to open source data (such as open source libraries used, library versions, dependencies between libraries, etc.), use clustering algorithms to cluster these features, classify program construction records that use similar open source data or have similar dependencies into one category, and form a second clustering result. The first clustering result and the second clustering result are fused. When fusion, it is necessary to consider factors such as the degree of overlap of the two clustering results, the distance between cluster centers, and the similarity within the cluster. The specific fusion can be determined by weighted fusion and other methods to determine the final program clustering result. Furthermore, the fused program clustering results are traversed, and the types within each cluster are identified. The program construction type of the cluster is determined based on the characteristics of the program construction records within the cluster (such as demand characteristics, open source data characteristics, etc.). The program construction type can be a specific classification label (such as Web application, data analysis tool, game, etc.) or a more fine-grained description (such as Web application based on a specific framework, data analysis tool using a specific algorithm, etc.).

[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, the open source file library and the multiple open source sub-libraries it contains are all extensible and updateable, which means that these libraries will be updated, add new features or fix problems over time. When dealing with these extensible and updateable libraries, it is necessary to ensure that the system or process can flexibly adapt to these changes. Optionally, for each open source library and sub-library, track its latest version and change log, and clearly specify the version of the library and sub-library used in the project to ensure consistency. Integrate dependency management tools into the project build process to automatically download and install the required library and sub-library versions at build time, and set up regular tasks or scripts to automatically check whether new library or sub-library versions have been released and evaluate whether they need to be updated. Before integrating a new version into the project, conduct sufficient testing to ensure compatibility, and ensure that there is a mechanism to roll back to the old version if the new version introduces problems or does not meet expectations. When the open source library or sub-library is updated, analyze whether these changes affect the recommendations and configurations of the build boot model, and based on the analysis results, update the build boot model to reflect the latest library and sub-library versions and possible best practices. In addition, there can be user notifications, documentation and maintenance, security considerations, and backup and recovery considerations.

[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] Exemplarily, by splitting a large open source library into multiple independent sub-libraries, a more refined modular design can be achieved. 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 updates 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 usually contain a large amount of data and business logic, which are the basis for the operation and function of the sub-database. Therefore, the maintenance and update of the data in the database is crucial. With the changes in business needs and the advancement of technology, the data in the database may need to be adjusted and optimized accordingly to ensure that the sub-database can continue to provide efficient, accurate and reliable services. The expansion and update of the data in the database can include data model updates, data processing optimization, and data migration and synchronization. For example, when business needs change, the data model in the sub-database may need to be updated to adapt to the new data structure or business logic, such as adding new fields, modifying field types, or adjusting data relationships; or in order to improve the performance and response speed of the sub-database, developers may need to optimize the data processing algorithm, including using more efficient data structures, improving algorithm logic, or introducing parallel processing technologies; or when the sub-database needs to be migrated to a new storage system or exchange data with other systems, data migration and synchronization operations may be required, such as migrating data from the old system to the new system and ensuring that the data remains consistent and accurate between multiple systems.

[0044] Extension and update based on sub-library division and extension and update based on data in the library are two important aspects of open source sub-library maintenance and update. The former focuses on the architecture and organization of the entire system, and improves the flexibility and scalability of the system through modular design; the latter focuses on the data and business logic within the sub-library, and ensures that the sub-library can continue to provide efficient, accurate and reliable services through data-driven and continuous iteration. In actual applications, developers can choose the appropriate extension and update method according to specific needs and situations 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:

[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] Based on the first program logic, build a first boot branch;

[0048] Identify the program building record and determine 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] Further, the system traverses all possible program building types, which are distinguished based on different programming languages, frameworks, design patterns or business logics. During the traversal process, a specific first type is extracted, which is selected based on a specific requirement, scenario or condition. For the extracted first type, the system analyzes the response results and program structure of the program under this type, and determines the first program logic by analyzing these results and structures. The logic involves a series of steps, conditional judgments, and function calls, which are used to achieve a specific function or business goal. Then, using the determined first program logic, the system starts to build a first guide branch, which may be a component of a user interface, a part of a decision tree, or a specific module of an AI model. The main function of the guide branch is to guide or guide users, developers or other systems to operate or make decisions according to the first program logic. After the first guide branch is built, program building records related to the first type are identified, which contain historical data, user feedback, test cases, etc. Then, data closely related to the first type is selected from these records as the first training sample. Next, the first training sample is used to supervise the training of the first guidance branch. During the training process, the system will evaluate the performance of the model and continuously adjust the parameters of the model to improve its accuracy. This process will continue until the model reaches a preset convergence. When the first guidance branch meets the preset convergence, it is considered to be built. At this time, 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 operate 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 can eventually build an accurate, stable, and practical guide branch to provide developers with efficient and standardized program building support.

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

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

[0054] The first guiding branch 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 guide branch is completed based on the same logic as the training of the first guide branch. After all the guide branches are trained, these branches are configured in parallel, that is, these branches will run at the same time, but each handles a different program building type. Parallel configuration can ensure that the system can handle a variety of different program building requirements at the same time, improving the overall processing efficiency and response speed. Then, the parallel configured guide branches are integrated into a complete building guide model, which will be able to automatically select the corresponding guide branch for processing according to the input program building type. At the same time, the building guide model also needs to ensure the collaboration between the guide branches to ensure the consistency and accuracy of the entire model. In addition, in the process of generating the building guide model, it is necessary to ensure that each guide branch has its own independent calculation domain, that is, each branch will not interfere with each other during operation, and each handles its own data and logic. Independent calculation domain can ensure the stability and reliability of each guide branch, and avoid affecting the operation of the entire model due to the failure or error of a branch. In addition, after the generation of the building guide model is completed, it is necessary to conduct sufficient testing to ensure its accuracy and reliability, and optimize and adjust the model according to the test results to improve its performance and accuracy.

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

[0057] Evenly distribute the branch computing power of the building 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, determining the adjustment ratio;

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

[0060] Exemplarily, the computing power required for each boot branch to operate normally is evaluated, and the specific method can be determined by simulation operation, historical data or expert evaluation. After understanding the computing power requirements of each branch, the total computing power resources are allocated to each branch according to the principle of balance to ensure that each branch can obtain sufficient computing power to support its normal operation. Record the computing power allocation results obtained by each branch as a benchmark for subsequent adjustments. Further, the program construction records are analyzed in detail, especially the construction frequency of each degree of construction type, so that it can be understood which construction types are more popular and which are less used. Then, based on the construction frequency, the ratio of each construction type to other types is calculated, and the ratio will be used for subsequent adjustment of the computing power allocation results. Next, determine how to adjust the computing power allocation of each branch according to the calculated adjustment ratio. Generally speaking, more popular construction types (i.e., types with higher construction frequency) should obtain more computing power support. The computing power of the initial allocation result is adjusted according to the adjustment strategy. Specifically, it can be achieved by increasing or decreasing the computing power quota of a specific branch. After the adjustment, a series of verification tests are performed to ensure that the new computing power allocation can meet the needs of various construction types and will not cause some branches to perform poorly due to insufficient computing power.

[0061] Through the above steps, the balanced distribution of computing power for building the boot model branches and adjustments based on actual needs can be achieved, which improves the overall performance and stability of the boot model and ensures that key or commonly used branches can obtain sufficient computing power resources to meet actual needs. At the same time, this adjustment method based on actual usage data also makes the entire boot model more in line with the needs of actual development scenarios, improving its practicality and effectiveness in actual applications.

[0062] Furthermore, the system further comprises a learning update module, wherein the learning update module is used to:

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

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

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

[0066] Specifically, define user behaviors, that is, clarify which operational behaviors are different from the behaviors of program construction guided by the model. These behaviors may be additional operations or modifications performed by users independently and independent of model guidance. Then, implement a mechanism to track and record these user behaviors. Specifically, this can be done through logging, event tracking, or a dedicated user behavior analysis system. And ensure that the collected user behavior data is comprehensive, accurate, and analyzable. Then, establish user behavior standards, that is, clarify what is the behavior that meets the construction standards. This may need to be formulated 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, mark the behaviors that meet the standards as valid learning behaviors, and do not meet the behaviors. The subsequent learning update process is not included.

[0067] Furthermore, a preset cycle, such as daily, weekly or monthly, is determined for regular identification and analysis of effective learning behaviors. At the end of each cycle, effective learning behaviors are screened out from the recorded user behavior data, and the screened effective learning behaviors are deeply analyzed to understand their patterns, trends and possible impacts. Based on the analysis results of effective learning behaviors, an update strategy for the building guidance model is formulated, including adjusting model parameters, optimizing algorithms, adding new features or improving the user interface. Use effective learning behaviors as training data to train the building guidance model. After the training is completed, the updated model is verified and tested to ensure that its performance and accuracy meet the expected standards. The verified updated model is deployed to the actual environment for user use and testing.

[0068] In summary, recording user behavior, making standard judgments on user behavior, and learning and updating the guidance model based on effective learning behavior can not only improve user experience and satisfaction, but also optimize the accuracy and applicability of the guidance model, realize self-iteration and optimization of the model, promote communication and knowledge sharing, and enhance the platform's adaptability and scalability. These effects jointly promote the platform to develop in a more intelligent and personalized direction.

[0069] Through the technical solutions of the above embodiments, the program building system based on user needs provided by the present application solves the technical problems in the prior art that the software has a low match 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 implementation manner does not constitute a limitation to the protection scope of the present 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 substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does 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, used 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, used to traverse the program building type, determine the underlying logic of the program, and construct a building guidance model based on the underlying logic of the program, wherein the building guidance model includes a plurality of guidance branches, and the plurality of guidance branches correspond to the program building type one by one; A channel establishment module, used to establish a direct calling channel between the building guidance model and the plurality of open source sub-libraries; The program generation module 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.

2. The program building system based on user needs as claimed in claim 1, characterized in that: The cluster segmentation module is also used for: Based on a preset period, reading the program building record of the application platform; Traversing the program construction records, performing a clustering based on the demand, determining a first clustering result, and performing a secondary clustering based on the open source data, determining a second clustering result; fusing the first clustering result and the second clustering result to determine a program clustering result; The program clustering results are traversed to identify the types within the cluster and determine the program building type.

3. The program building system based on user needs as claimed in claim 1, characterized in that: The open source file library is extensible and updatable, and the plurality of open source sub-libraries are extensible and updatable.

4. The program building system based on user needs as claimed in claim 3, 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.

5. The program building system based on user needs as claimed in claim 2, 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; Based on the first program logic, build a first boot branch; Identify the program building record and determine 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.

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

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

8. The program building system based on user needs as claimed in claim 1, characterized in that: The system further comprises a learning update module, wherein the learning update module is used for: Recording user behavior, wherein the user behavior refers to the operation behavior different from the program construction guided by the model; Conduct a standard determination on the user behavior, and if it meets the construction standard, it is regarded as a valid learning behavior; Based on a preset cycle, the effective learning behavior is identified, and the building guidance model is updated through learning.

Images