Integrated development method and device for small programs

Abstract

The embodiment of the present application provides an integrated development method and device for mini-programs, the method comprising: receiving a mini-program development instruction sent by a user, determining a corresponding device type and operating system environment according to the mini-program development instruction, and determining a corresponding mini-program function summary feature according to the device type and operating system environment; inputting the decision tree prediction feature into a feature integration layer of a preset regression tree model, determining the corresponding mini-program function feature according to the decision tree prediction feature, the regression tree prediction feature output by the pruning layer in the preset regression tree model, and a preset model integration algorithm, determining the corresponding mini-program function feature according to the mini-program function feature, preset real features, and a back propagation algorithm, and performing a mini-program project generation operation according to the mini-program function module model; the present application can perform integrated mini-program development in a more intelligent, accurate, and efficient manner to improve development efficiency and mini-program quality.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an integrated development method and device for mini-programs. Background Art

[0002] In the current mobile Internet and digital application environment, mini programs have become an important choice for many companies and developers due to their lightweight, efficient, and cross-platform features. However, with the increasing demand for mini program functions and the improvement of user experience, how to achieve efficient, intelligent, and integrated mini program development has become an urgent problem to be solved. Although existing technologies have made certain progress in the process of mini program development, there are still many challenges and shortcomings in responding to diversified needs, cross-platform compatibility, development efficiency, and intelligence.

[0003] First, existing mini-program development methods usually rely on developers to manually write code and configure the environment. This development model is not only inefficient, but also prone to errors. Especially when facing multiple device types and different operating system environments, developers need to perform separate adaptation and debugging for each situation, which consumes a lot of time and energy. In addition, in the process of manual coding and configuration, developers need to have a high level of technical skills. For inexperienced developers, development is difficult and the learning cost is high.

[0004] Secondly, in the existing mini-program development process, the extraction and analysis of functional features mainly rely on the experience and intuition of developers. This method is not only highly subjective, but also easily misses important functional features, resulting in the mini-programs developed being unable to fully meet user needs. With the increasing complexity and diversity of mini-program functions, methods that rely solely on manual experience are obviously difficult to meet demand. Even if developers have rich experience, it is difficult to accurately extract and analyze all functional features in a short period of time, affecting development efficiency and mini-program quality.

[0005] In terms of model selection and optimization, existing mini-program development methods usually use traditional development models and tools, such as the MVC (Model-View-Controller) model and the MVVM (Model-View-ViewModel) model. Although these development models can improve development efficiency and code maintainability to a certain extent, they still have performance bottlenecks when dealing with complex functions and cross-platform compatibility. In particular, when dealing with high-dimensional data and complex functions, traditional development models are prone to performance degradation or extended development cycles, affecting the timeliness of mini-program launch.

[0006] In addition, the existing mini-program development system lacks effective intelligent development auxiliary tools. Although some development platforms provide functions such as automatic code generation and template reuse, the intelligence level and application scope of these tools still need to be improved in actual applications. Especially when facing multi-source heterogeneous data and complex functional requirements, how to effectively use intelligent tools to improve development efficiency and mini-program quality is an urgent problem to be solved.

[0007] In terms of feedback and optimization of development results, existing technologies are also insufficient. Most existing systems rely on manual testing and debugging by developers and lack the ability to dynamically adjust and self-optimize. Due to the constant changes in user needs and market environment, fixed development strategies are difficult to maintain long-term effectiveness. In addition, during the project generation process, existing systems often ignore user feedback on development results and lack effective feedback mechanisms and model adjustment methods, which makes it difficult to continuously optimize development results and affects user experience.

[0008] In summary, the existing mini-program development methods still have many deficiencies and challenges in terms of user demand capture, functional feature extraction, model selection and optimization, development result feedback and adjustment, etc. Faced with increasingly complex user needs and market environment, a more intelligent, accurate and efficient mini-program integrated development method is urgently needed to improve development efficiency and mini-program quality. Summary of the invention

[0009] In response to the problems in the prior art, the present application provides an integrated development method and device for mini-programs, which can perform integrated development of mini-programs in a more intelligent, accurate and efficient manner, so as to improve development efficiency and mini-program quality.

[0010] In order to solve at least one of the above problems, the present application provides the following technical solutions:

[0011] In a first aspect, the present application provides an integrated development method for mini-programs, including:

[0012] Receive a mini-program development instruction sent by a user, determine a device type and an operating system environment corresponding to the mini-program development instruction, and determine a corresponding mini-program function summary feature according to the device type and the operating system environment;

[0013] Input the mini-program function summary features into the feature selection layer of a preset decision tree model, determine the corresponding decision tree selection features according to the preset segmentation rules, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operations, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0014] The decision tree prediction feature is input into the feature integration layer of the preset regression tree model, and the corresponding mini-program function feature is determined according to the decision tree prediction feature, the regression tree prediction feature output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, and the corresponding mini-program function module model is determined according to the mini-program function feature, the preset real feature, and the back propagation algorithm, and the mini-program project generation operation is performed according to the mini-program function module model.

[0015] Furthermore, the determining of the corresponding mini-program function summary features according to the device type and the operating system environment includes:

[0016] Performing a minority class identification operation on the device type and the operating system environment according to a preset minority class identification rule, and determining corresponding nearest neighbor mini-program functional features according to the minority class mini-program functional features obtained after the minority class identification operation and a preset K nearest neighbor algorithm;

[0017] The corresponding mini-program function summary features are determined according to the minority mini-program function features and the nearest neighbor mini-program function features.

[0018] Furthermore, the determining of the corresponding nearest neighbor mini-program functional features based on the minority mini-program functional features obtained after the minority class identification operation and a preset K nearest neighbor algorithm includes:

[0019] Performing a distance calculation operation on the minority mini-program functional features and the nearest neighbor mini-program functional features to determine the corresponding mini-program functional feature distances;

[0020] The corresponding synthesized mini-program functional features are determined according to the mini-program functional feature distance and a preset ratio rule, and the mini-program functional features are subjected to feature addition operations according to the synthesized mini-program functional features to determine the corresponding mini-program functional summary features.

[0021] Furthermore, the step of inputting the mini-program function summary features into a feature selection layer of a preset decision tree model and determining corresponding decision tree selection features according to preset segmentation rules includes:

[0022] Input the mini program function summary features into the feature selection layer of a preset decision tree model, and determine the corresponding feature segmentation points according to the preset Gini impurity;

[0023] Determine the corresponding decision tree selection features according to the feature segmentation points and the mini-program function summary features.

[0024] Furthermore, the step of inputting the decision tree selection feature into a splitting layer of a preset decision tree model to perform a recursive splitting operation, determining a corresponding decision tree recursive feature, and determining a decision tree prediction feature output by a corresponding preset decision tree model according to the decision tree recursive feature comprises:

[0025] Inputting the decision tree selection features into the splitting layer of a preset decision tree model, performing a recursive splitting operation according to a preset maximum tree depth, and determining corresponding decision tree recursive features;

[0026] The decision tree recursive features are pruned to determine the decision tree prediction features output by the corresponding preset decision tree model, wherein the decision tree prediction features are used to predict category labels.

[0027] Furthermore, before inputting the decision tree prediction features into the feature integration layer of the preset regression tree model, the method includes:

[0028] Input the mini-program function summary features into the feature selection layer of a preset regression tree model, perform feature selection operations according to a preset mean square error algorithm, determine corresponding regression tree selection features, perform recursive splitting operations according to the regression tree selection features, and determine corresponding regression tree recursive features;

[0029] The regression tree recursive features are input into the pruning layer of the preset regression tree model, the regression tree recursive features are pruned, the corresponding regression tree prediction features are determined, and the regression tree prediction features are input into the feature integration layer of the preset regression tree model, wherein the regression tree prediction features are used to predict continuous values.

[0030] Furthermore, the determining of the corresponding mini-program functional features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm includes:

[0031] Perform a weight matching operation based on the decision tree prediction feature and the regression tree prediction feature output by the pruning layer in the preset regression tree model to determine the corresponding decision tree prediction feature weight and regression tree prediction feature weight;

[0032] A feature weighted averaging operation is performed based on the decision tree prediction feature weights and the regression tree prediction feature weights to determine the corresponding mini-program functional features.

[0033] In a second aspect, the present application provides an integrated development device for mini-programs, including:

[0034] A function feature collection module, used to receive a mini-program development instruction sent by a user, determine a corresponding device type and operating system environment according to the mini-program development instruction, and determine a corresponding mini-program function summary feature according to the device type and operating system environment;

[0035] A function feature prediction module, used to input the mini-program function summary features into the feature selection layer of a preset decision tree model, determine the corresponding decision tree selection features according to a preset segmentation rule, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0036] The mini program construction module is used to input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model and the preset model integration algorithm, determine the corresponding mini program push model according to the mini program function features, preset real features and the back propagation algorithm, and perform the mini program push operation according to the mini program push model.

[0037] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the integrated development method for mini-programs when executing the program.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the integrated development method for mini-programs.

[0039] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the integrated development method for mini-programs.

[0040] It can be seen from the above technical scheme that the present application provides an integrated development method and device for mini-programs, which receives mini-program development instructions sent by users, determines the corresponding device type and operating system environment according to the mini-program development instructions, and determines the corresponding mini-program function summary characteristics according to the device type and operating system environment; inputs the decision tree prediction features into the feature integration layer of the preset regression tree model, determines the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determines the corresponding mini-program function features according to the mini-program function features, preset real features, and the back propagation algorithm, and performs the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient integrated development of mini-programs to improve development efficiency and mini-program quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 This is one of the flowcharts of the integrated development method for mini-programs in the embodiment of the present application;

[0043] Figure 2 This is the second flow chart of the integrated development method for mini-programs in the embodiment of the present application;

[0044] Figure 3 The third flowchart of the integrated development method for mini-programs in the embodiment of the present application;

[0045] Figure 4 This is a fourth flow chart of the integrated development method for mini-programs in an embodiment of the present application;

[0046] Figure 5 This is the fifth flow chart of the integrated development method for mini-programs in the embodiment of the present application;

[0047] Figure 6 This is the sixth flow chart of the integrated development method for mini-programs in the embodiment of the present application;

[0048] Figure 7 This is the seventh flow chart of the integrated development method for mini-programs in the embodiment of the present application;

[0049] Figure 8 This is a structural diagram of an integrated development device for mini-programs in an embodiment of the present application;

[0050] Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0051] Reference numerals:

[0052] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0054] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0055] In view of the problems existing in the prior art, the present application provides an integrated development method and device for mini-programs, which receives mini-program development instructions sent by users, determines the corresponding device type and operating system environment according to the mini-program development instructions, and determines the corresponding mini-program function summary characteristics according to the device type and operating system environment; inputs the decision tree prediction features into the feature integration layer of the preset regression tree model, determines the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determines the corresponding mini-program function module model according to the mini-program function features, preset real features and the back propagation algorithm, and performs the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient mini-program integrated development.

[0056] In order to be able to develop mini-programs in an integrated manner more intelligently, accurately and efficiently, so as to improve development efficiency and mini-program quality, this application provides an embodiment of an integrated development method for mini-programs, see Figure 1 The integrated development method for mini-programs specifically includes the following contents:

[0057] Step S101: receiving a mini-program development instruction sent by a user, determining a corresponding device type and operating system environment according to the mini-program development instruction, and determining a corresponding mini-program function summary feature according to the device type and operating system environment;

[0058] Optionally, in this embodiment, in step S101, the system first needs to receive the mini program development instruction sent by the user, and determine the corresponding device type and operating system environment according to the instruction. Then, the system will use this information to determine the mini program function summary features, thereby laying the foundation for subsequent mini program development. The core of this step is to accurately analyze the user's needs and generate suitable function summary features according to the device type and operating system environment.

[0059] The specific technical process is that the system receives the mini program development instructions sent by the user through the interface. The instructions usually contain the user's basic requirements for the mini program, such as application scenarios, target user groups, and functional requirements. The system parses the instruction content and extracts key information, such as device type (such as smartphones, tablets, smart watches, etc.) and operating system environment (such as Android, iOS, HarmonyOS, etc.). During the parsing process, the system will use natural language processing technology to analyze the instruction content to ensure that the key information in the user's needs can be accurately extracted.

[0060] In terms of technical principles, the system uses natural language processing (NLP) technology to parse user instructions. NLP technology extracts key information such as device type and operating system environment from user instructions through grammatical and semantic analysis of text content. Then, the system accesses the pre-established feature library based on this information to match the corresponding mini-program function profile features. The feature library stores mini-program function feature templates for various device types and operating system environments. The system generates function profile features that meet user needs through the matching process.

[0061] In terms of the technical problems solved, step S101 mainly solves the problem of how to accurately parse user needs and generate mini-program function summary features according to different device types and operating system environments. Different devices and operating systems may have different functional requirements and implementation methods for mini-programs, and directly using general templates may not meet the special needs of users. By parsing user instructions and matching templates in the feature library, the system can generate more targeted and adaptive mini-program function summary features, improving development efficiency and user satisfaction.

[0062] In terms of the technical effects achieved, through step S101, the system can accurately receive and parse the mini-program development instructions sent by the user, and generate suitable mini-program function profile features according to the device type and operating system environment. In this way, the system not only improves the efficiency of mini-program development, but also ensures that the developed mini-program can better adapt to the target device and operating system environment, providing a better user experience.

[0063] For example, suppose a user wants to develop a fitness tracking applet for smart watches and sends a command: "Please help me develop a fitness tracking applet that can be used on smart watches." After receiving the command, the system first uses NLP technology to parse out the key information: "smart watch" and "fitness tracking." Then, the system searches the feature library for the summary features of fitness tracking functions suitable for smart watches, such as exercise data recording, heart rate monitoring, step counting, etc.

[0064] After parsing the device type and functional requirements, the system will further refine the functional profile features based on the operating system environment of the smartwatch (such as Wear OS, watchOS, etc.). For example, if the target device is a smartwatch using Wear OS, the system will add APIs and interaction design specifications compatible with Wear OS to the functional profile features. In this way, the system generates a detailed and adaptive mini-program functional profile feature, including fitness data recording, heart rate monitoring, step counting, calorie consumption calculation and other functions.

[0065] Finally, based on these functional profile features, the development team can quickly enter the development phase to ensure that the developed mini-program not only meets the functional requirements of users, but also runs smoothly on smart watches and provides a good user experience. Through this refined demand analysis and functional feature generation process, the system can significantly improve the efficiency and quality of mini-program development and meet the personalized needs of users.

[0066] Step S102: Input the mini-program function summary features into the feature selection layer of the preset decision tree model, determine the corresponding decision tree selection features according to the preset segmentation rules, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0067] Optionally, in this embodiment, in step S102, the system needs to input the determined mini-program function summary features into the feature selection layer of the preset decision tree model. The corresponding decision tree selection features are determined by the preset segmentation rules, and then these features are input into the splitting layer for recursive splitting operations, and finally the decision tree recursive features are determined, and decision tree prediction features are generated based on these features. The core of this step is to use the decision tree model to process and optimize the function summary features, thereby generating prediction features that help improve the efficiency and quality of mini-program development.

[0068] The specific technical process is that the system first inputs the mini program function summary features generated in step S101 into the feature selection layer of the preset decision tree model. In the feature selection layer, the system will filter and segment the input features according to the preset segmentation rules. The segmentation rules are usually set based on indicators such as feature importance and information gain. These rules ensure that the selected features can reflect the core functional requirements and optimization direction of the mini program to the greatest extent.

[0069] Next, the system inputs the selected decision tree selection features into the splitting layer of the decision tree model. In the splitting layer, the system further refines and optimizes the features through recursive splitting operations. Recursive splitting operations mean that the system starts from the root node and continuously splits the features into smaller sub-nodes according to the distribution and segmentation rules of the features until the preset stop conditions (such as node purity, maximum depth, etc.) are reached. In this process, the system will evaluate the effect of each splitting point and select the optimal splitting method to ensure that the generated sub-nodes can maximize the prediction effect of the model.

[0070] In terms of technical principles, the operations of the feature selection layer and the splitting layer are based on the core algorithm of the decision tree model. The feature selection layer uses indicators such as information gain to screen and split features to ensure that the selected features can effectively reflect the core information of the data. The splitting layer continuously optimizes the distribution and segmentation of features through recursive splitting operations to generate a layer-by-layer feature tree structure. This process not only improves the expressiveness of features, but also enhances the prediction effect and stability of the model.

[0071] In terms of the technical problems to be solved, step S102 mainly solves the problem of how to use the decision tree model to process and optimize the mini-program function summary features. Mini-program function summary features usually contain a lot of redundant information and noise, and directly using these features for development may lead to inefficiency and poor quality. Through feature selection and recursive splitting operations, the system can filter out the most important features, optimize the distribution and structure of features, and improve the prediction effect of the model and the development efficiency of the mini-program.

[0072] In terms of the technical effects achieved, through step S102, the system can input the mini-program function summary features into the decision tree model, and generate optimized decision tree prediction features through feature selection and recursive splitting operations. These prediction features not only accurately reflect the core functional requirements of the mini-program, but also effectively guide subsequent development work and improve the development efficiency and quality of the mini-program. Ultimately, the system can generate efficient, stable mini-programs with high prediction effects, providing users with high-quality development services.

[0073] For example, suppose a user wants to develop a fitness tracking applet for a smartwatch and sends a command: "Please help me develop a fitness tracking applet that can be used on a smartwatch." In step S101, the system has generated functional summary features including exercise data recording, heart rate monitoring, step counting, etc. In step S102, the system inputs these functional summary features into the feature selection layer of the preset decision tree model.

[0074] In the feature selection layer, the system selects the most important features based on preset segmentation rules (such as information gain). For example, the system may find that "heart rate monitoring" and "step counting" are the core functions that users are most concerned about, so these features are used as decision tree selection features. Then, the system inputs these selected features into the splitting layer for recursive splitting operations. Through recursive splitting, the system continuously optimizes the distribution and structure of features to generate a layer-by-layer feature tree structure. For example, the system may further refine the "heart rate monitoring" function and split out sub-features such as "real-time heart rate monitoring" and "abnormal heart rate alarm".

[0075] Finally, the system determines the recursive features of the decision tree based on the feature tree structure generated by recursive splitting, and generates the prediction features of the decision tree. These prediction features not only accurately reflect the user's core functional requirements for the mini program, but also effectively guide the work of the development team. For example, the development team can focus on developing and optimizing functions such as "real-time heart rate monitoring" and "abnormal heart rate alarm" based on the prediction features to ensure that the mini program can run smoothly on the smart watch and provide a high-quality user experience.

[0076] In this way, the system can significantly improve the efficiency and quality of mini-program development, meet the personalized needs of users, and provide users with efficient, stable and highly predictive mini-program development services.

[0077] Step S103: Input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determine the corresponding mini-program function module model according to the mini-program function features, preset real features, and the back propagation algorithm, and perform the mini-program project generation operation according to the mini-program function module model.

[0078] Optionally, in this embodiment, in step S103, the system needs to input the decision tree prediction features obtained in step S102 into the feature integration layer of the preset regression tree model. Based on these decision tree prediction features, and the regression tree prediction features output by the pruning layer in the regression tree model, combined with the preset model integration algorithm, the corresponding mini-program functional features are determined. Next, the system will use these functional features, the preset real features and the back propagation algorithm to determine the corresponding mini-program functional module model, and perform the mini-program project generation operation based on the model.

[0079] The specific technical process is to first input the decision tree prediction features into the feature integration layer of the regression tree model. The main task of the feature integration layer is to integrate and weight the input features to obtain a more accurate and optimized feature set. In the feature integration layer, the system will combine the regression tree prediction features output by the pruning layer, and comprehensively evaluate and optimize these features through model integration algorithms (such as weighted averaging, Boosting, etc.), thereby generating accurate mini program functional features.

[0080] Next, the system uses these mini-program functional features and the preset real features to determine the mini-program functional module model through the back propagation algorithm. The back propagation algorithm is a common machine learning optimization algorithm that continuously adjusts the model parameters to make the model output as close to the real features as possible. In this process, the system will continuously iterate and optimize the model parameters until the model's prediction results meet the preset accuracy requirements.

[0081] In terms of technical principles, the operation of the feature integration layer is based on model integration algorithms, which improve the overall prediction performance by integrating the prediction results of multiple models. The regression tree prediction features of the pruning layer are pruned to remove redundant and noisy features and retain the most important features, thereby improving the accuracy and stability of the model. The back propagation algorithm continuously adjusts the model parameters through optimization methods such as gradient descent, so that the output of the model is more consistent with the real features.

[0082] In terms of the technical problems to be solved, step S103 mainly solves the problem of how to effectively integrate features from multiple sources, generate accurate mini-program functional features, and build and optimize the mini-program functional module model based on these features. In practical applications, different feature sources may contain redundant information and noise, and directly using these features for modeling may lead to a decrease in model accuracy. Through feature integration and pruning operations, the system can remove redundant and noisy features and retain important features, thereby improving the prediction performance and stability of the model.

[0083] In terms of the technical effects achieved, through step S103, the system can effectively integrate the decision tree prediction features and the regression tree prediction features to generate accurate mini-program functional features. These functional features not only accurately reflect the core requirements of the mini-program, but also effectively guide the development and optimization of functional modules. Through the back propagation algorithm, the system can build and optimize the mini-program functional module model to ensure that the model's prediction results are highly consistent with the actual features. Finally, the system can efficiently generate mini-program projects based on the optimized functional module model to meet the user's development needs.

[0084] For example, suppose a user wants to develop a fitness tracking applet for smart watches. In step S102, the system has generated decision tree prediction features including "real-time heart rate monitoring" and "step count". In step S103, the system inputs these prediction features into the feature integration layer of the regression tree model, and generates comprehensive applet functional features by combining the regression tree prediction features such as "calorie consumption calculation" and "exercise data recording" output by the pruning layer and using the model integration algorithm (such as weighted average).

[0085] Then, the system uses these functional features and preset real features (such as actual user exercise data) to build and optimize the mini-program functional module model through the back-propagation algorithm. In this process, the system will continuously adjust the model parameters so that the output of the model (such as the accuracy of heart rate monitoring and the accuracy of step count) is highly consistent with the real features (actually measured heart rate and step count). After multiple iterations and optimizations, the system finally generates an accurate mini-program functional module model.

[0086] Based on this optimized functional module model, the system can efficiently generate mini program projects. For example, the development team can focus on optimizing real-time heart rate monitoring and step counting functions based on the output of the model to ensure that the mini program can provide accurate fitness data tracking and monitoring services on smart watches.

[0087] From the above description, it can be seen that the integrated development method for mini-programs provided in the embodiment of the present application can receive mini-program development instructions sent by users, determine the corresponding device type and operating system environment according to the mini-program development instructions, and determine the corresponding mini-program function summary characteristics according to the device type and operating system environment; input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determine the corresponding mini-program function features according to the mini-program function features, preset real features and the back propagation algorithm, and perform the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient mini-program integrated development.

[0088] In one embodiment of the integrated development method for mini-programs of the present application, see Figure 2 , and can also include the following:

[0089] Step S201: performing a minority class identification operation on the device type and the operating system environment according to a preset minority class identification rule, and determining corresponding nearest neighbor mini-program functional features according to the minority class mini-program functional features obtained after the minority class identification operation and a preset K nearest neighbor algorithm;

[0090] Step S202: Determine the corresponding mini-program function summary features according to the minority mini-program function features and the nearest neighbor mini-program function features.

[0091] Optionally, in this embodiment, in step S201, the system first performs minority class recognition operations on the device type and operating system environment according to preset minority class recognition rules. The purpose of minority class recognition is to identify those uncommon or small sample categories in the data set so that special attention can be given to them in subsequent processing to avoid these minority classes being ignored or misjudged in the model. This process is particularly important because minority class data may contain key functional requirements or special operating environments in the development of mini-programs, which directly affects the user experience and function implementation of the mini-programs.

[0092] The specific technical process is that the system first analyzes and classifies the data of device type and operating system environment according to the preset minority class identification rules. Minority class identification rules usually include statistical analysis methods, such as sample quantity statistics, density estimation, etc., to determine which categories belong to the minority class. For example, in a database containing a variety of smart devices and operating systems, the system may find that some old devices or specific versions of operating systems have a small number of samples, thereby identifying them as minority classes.

[0093] Once the minority classes are identified, the system extracts the mini-program functional features corresponding to these categories, which are called minority mini-program functional features. These features may contain functional requirements or operational limitations that are unique to these minority devices and operating systems. Next, the system uses the preset K nearest neighbor algorithm to determine the corresponding nearest neighbor mini-program functional features based on the minority mini-program functional features. The K nearest neighbor algorithm is a distance-based classification algorithm that calculates the distance between features to find the feature set that is most similar to the minority class features, thereby ensuring that the importance of minority class features is reflected in the model.

[0094] In terms of technical principles, minority class identification is based on statistical analysis and machine learning methods. By analyzing the data distribution, it identifies categories with fewer samples but higher importance. The K nearest neighbor algorithm calculates the distance in the feature space to find the features most similar to the minority class features, ensuring that the importance of these features is retained and strengthened in subsequent processing.

[0095] In terms of the technical problems solved, step S201 mainly solves the problem of how to identify and process minority class features in the data set. Minority class features are easily ignored or misjudged in traditional algorithms, resulting in poor performance of the model when processing these categories. Through minority class identification and K nearest neighbor algorithm, the system can ensure that minority class features are fully valued and optimized in the model, thereby improving the performance and accuracy of the model when processing these features.

[0096] In terms of the technical effects achieved, through step S201, the system can effectively identify and process minority class features, ensuring that these features are fully valued and optimized in the model. The nearest neighbor mini program functional features finally generated can not only accurately reflect the needs of minority class features, but also effectively guide the subsequent mini program development work, improving the functional coverage and user experience of the mini program.

[0097] For example, suppose a user wants to develop a fitness tracking app that is suitable for multiple devices and operating systems. In the dataset, the system finds that some old devices (such as old smart watches) and specific versions of operating systems (such as early Android versions) have fewer samples, but these devices and operating systems still have a certain number of users. The system uses minority class identification rules to identify these devices and operating systems as minority classes.

[0098] Next, the system extracts the mini-program functional features corresponding to these minority devices and operating systems, such as specific sensor data formats, low memory requirements, etc. Then, the system uses the K nearest neighbor algorithm to find the mini-program functional features that are most similar to these minority features and generate the nearest neighbor mini-program functional features. For example, the system may find that the features of some new devices are similar to those of these old devices, and thus include them in the nearest neighbor feature set.

[0099] In step S202, the system determines the corresponding mini-program function summary features based on the minority mini-program function features and the nearest neighbor mini-program function features. The purpose of this step is to integrate the minority and nearest neighbor features to generate a comprehensive mini-program function summary feature set. The specific technical process is that the system merges and optimizes the minority features and the nearest neighbor features through a feature fusion algorithm to generate a comprehensive mini-program function summary feature.

[0100] In terms of technical principles, the feature fusion algorithm effectively integrates minority features and nearest neighbor features through weighted averaging, feature superposition and other methods, ensuring that the generated functional summary features can fully reflect the functional requirements and operating environment of the mini program. In this way, the system can fully consider the impact of minority features during the development process and improve the functional coverage and user experience of the mini program.

[0101] In terms of the technical problems to be solved, step S202 mainly solves the problem of how to effectively integrate minority class features and nearest neighbor features. Minority class features and nearest neighbor features may be processed separately in traditional algorithms, resulting in the integrity and consistency of functional features being affected. Through the feature fusion algorithm, the system can ensure that the generated mini program functional summary features are comprehensive and consistent, thereby improving the development quality and user satisfaction of the mini program.

[0102] In terms of the technical effects achieved, through step S202, the system can generate a comprehensive and optimized set of mini-program function summary features. These features can not only accurately reflect the special needs of a small number of devices and operating systems, but also effectively guide the work of the development team and improve the function coverage and user experience of the mini-program. Ultimately, the system can efficiently generate mini-program projects based on these optimized function summary features to meet the user's usage needs on a variety of devices and operating systems.

[0103] In general, through steps S201 and S202, the system can effectively identify and process minority class features, ensuring that these features are fully valued and optimized in the model.

[0104] In one embodiment of the integrated development method for mini-programs of the present application, see Figure 3 , and can also include the following:

[0105] Step S301: performing a distance calculation operation on the minority mini-program functional features and the nearest neighbor mini-program functional features to determine the corresponding mini-program functional feature distances;

[0106] Step S302: Determine the corresponding synthesized mini-program functional features according to the mini-program functional feature distance and a preset ratio rule, perform feature adding operations on the mini-program functional features according to the synthesized mini-program functional features, and determine the corresponding mini-program functional summary features.

[0107] Optionally, in this embodiment, in step S301, the system first performs a distance calculation operation on the minority class mini-program functional features and the nearest neighbor mini-program functional features to determine the corresponding mini-program functional feature distances. Specifically, the distance calculation operation measures the similarity between the minority class features and the nearest neighbor features by a certain distance measurement method (such as Euclidean distance, Manhattan distance, etc.). The purpose of this step is to quantify the difference between the minority class features and their nearest neighbor features, thereby providing a basis for subsequent feature fusion.

[0108] The specific technical process is that the system first extracts the feature vectors of the minority class mini-program function features and the nearest neighbor mini-program function features, and then uses the preset distance measurement method to calculate the distance between them. For example, for two feature vectors, the system can calculate their Euclidean distance in the multidimensional feature space. This distance value reflects the similarity of the two feature vectors in the feature space. The smaller the distance, the more similar the two features are.

[0109] In terms of technical principles, distance calculation is based on the vector space model, using mathematical methods to operate on feature vectors. Common distance measurement methods include Euclidean distance, Manhattan distance, and cosine similarity, which measure the similarity or difference between feature vectors through different calculation formulas. By calculating the distance between minority class features and nearest neighbor features, the system can quantify the relationship between these features, providing a basis for subsequent feature fusion and optimization.

[0110] In terms of the technical problems solved, step S301 mainly solves the problem of how to quantify the similarity between minority class features and nearest neighbor features. In a multidimensional feature space, simple feature comparison may not be able to accurately measure the differences between features. Through distance calculation, the system can accurately quantify these differences, thus providing a basis for subsequent feature fusion and optimization.

[0111] In terms of the technical effects achieved, through step S301, the system can accurately calculate the distance between the minority class features and the nearest neighbor features, providing a quantitative basis for subsequent feature fusion and optimization. These distance values ​​can not only reflect the similarity between features, but also guide the weight allocation and optimization strategy in the feature fusion process, thereby improving the accuracy and consistency of the mini program's functional features.

[0112] For example, suppose a user wants to develop a music player app that works on multiple devices and operating systems. In the dataset, the system identifies the audio processing features of some older devices and the compatibility features of specific operating systems as minority classes. The system measures the similarity between these minority class features and the features of new devices or mainstream operating systems through distance calculation. For example, the system can calculate the Euclidean distance between the audio processing features of older devices and the audio processing features of new smartphones to quantify the similarity between them.

[0113] In step S302, the system determines the corresponding synthetic mini-program functional features based on the mini-program functional feature distance and the preset ratio rule. The preset ratio rule refers to how to reasonably allocate the weights of minority class features and nearest neighbor features in the feature fusion process to generate synthetic features. The specific technical process is that the system first performs weighted processing on the minority class features and nearest neighbor features based on the distance calculation results to generate synthetic mini-program functional features. The weighted processing is usually based on a preset ratio rule, such as the smaller the distance, the higher the weight, to ensure that features with high similarity occupy a larger proportion in the synthetic features.

[0114] Next, the system adds features to the original mini-program features based on the synthesized mini-program features to determine the corresponding mini-program feature summary features. Feature addition refers to fusing and expanding the synthesized features with the original features to generate a more comprehensive mini-program feature set. This process not only takes into account the combined impact of minority features and nearest neighbor features, but also ensures that the final generated feature set can fully cover the mini-program's functional requirements and operating environment.

[0115] In terms of technical principles, feature fusion and weighted processing are based on weighted averaging and feature expansion methods, which integrate and optimize multiple features by reasonably allocating feature weights. The preset proportion rule dynamically adjusts the feature weight allocation based on the distance calculation results to ensure that the synthesized features can accurately reflect the combined impact of minority class features and nearest neighbor features.

[0116] In terms of the technical problems to be solved, step S302 mainly solves the problem of how to effectively fuse minority class features and nearest neighbor features. In traditional methods, simple feature fusion may not accurately reflect the similarities and differences between features. Through weighted processing and feature expansion, the system can generate a more accurate and comprehensive mini-program feature set.

[0117] In terms of the technical effects achieved, through step S302, the system can generate synthetic mini-program functional features, and through feature addition operations, generate comprehensive and optimized mini-program functional summary features. These features can not only accurately reflect the special needs of a small number of devices and operating systems, but also effectively guide the work of the development team and improve the functional coverage and user experience of the mini-program. Ultimately, the system can efficiently generate mini-program projects based on these optimized functional summary features to meet the user's usage needs on a variety of devices and operating systems.

[0118] In general, through steps S301 and S302, the system can effectively calculate the distance between the minority class features and the nearest neighbor features, and generate comprehensive and optimized mini program function summary features through weighted processing and feature expansion. This not only improves the accuracy and consistency of mini program function features, but also provides strong support and guidance for subsequent mini program development.

[0119] In one embodiment of the integrated development method for mini-programs of the present application, see Figure 4 , and can also include the following:

[0120] Step S401: input the mini-program function summary features into the feature selection layer of a preset decision tree model, and determine the corresponding feature segmentation points according to the preset Gini impurity;

[0121] Step S402: Determine corresponding decision tree selection features according to the feature segmentation points and the mini-program function summary features.

[0122] Optionally, in this embodiment, in step S401, the system first inputs the mini-program function summary features into the feature selection layer of a preset decision tree model. The decision tree model is a machine learning algorithm that generates tree-structured decision rules by segmenting and classifying data. Here, the role of the feature selection layer is to select the most important features that can best distinguish different categories or features based on the input function summary features. The specific technical process of implementation is that the system first takes the mini-program function summary features as input and passes them to the feature selection layer of the decision tree model.

[0123] The feature selection layer determines the corresponding feature split point by calculating the Gini impurity of each feature. Gini impurity is an indicator to measure the impurity of a data set. The smaller the value, the higher the purity of the data set. In the decision tree model, Gini impurity is used to select split nodes, and the feature that can minimize the impurity of the data set is selected for splitting. Specifically, the system traverses all possible split points for each feature, calculates the change in Gini impurity before and after the split, and selects the split point with the largest reduction in Gini impurity as the feature split point.

[0124] In terms of technical principles, Gini impurity measures the distinguishing ability of features by calculating the classification effect of features at different segmentation points. The decision tree model uses this principle to continuously select the segmentation point with the lowest Gini impurity, build the decision tree structure layer by layer, and finally generate a small program function feature decision tree that can effectively classify and predict.

[0125] In terms of the technical problems solved, step S401 mainly solves the problem of how to select the most discriminating features from among the many mini-program function summary features. Since mini-program function features may contain a large number of different types of features, simple feature selection methods may not be able to accurately find the most important features. By calculating the Gini impurity, the system can scientifically measure the discriminating power of each feature, thereby ensuring that the selected features can maximize the classification effect and accuracy of the model.

[0126] In terms of the technical effects achieved, through step S401, the system can effectively input the mini-program function summary features into the decision tree model, and determine the optimal feature segmentation point through Gini impurity calculation. This not only improves the accuracy of feature selection, but also lays a solid foundation for the subsequent decision tree construction. The feature segmentation points finally generated can accurately reflect the differences between the mini-program function features, providing strong support for subsequent feature selection and optimization.

[0127] For example, suppose a user wants to develop a health monitoring applet that is suitable for multiple devices and operating systems. The system collects functional features of different devices and operating systems, such as sensor data collection frequency, data processing capabilities, and user interaction methods. Through step S401, the system inputs these functional summary features into the feature selection layer of the preset decision tree model. The feature selection layer calculates the Gini impurity to find the best split point of the sensor data collection frequency on different devices, thereby selecting the most distinguishing feature split point.

[0128] In step S402, the system determines the corresponding decision tree selection features based on the feature segmentation points and the mini-program function summary features. The specific technical process is that the system first segments and classifies the mini-program function summary features based on the feature segmentation points determined in step S401. Then, the system constructs the selection features of the decision tree based on the segmented features to generate the final decision tree model.

[0129] In terms of technical principles, the decision tree selects features through recursive segmentation and classification, continuously selecting the segmentation point with the lowest Gini impurity, and building the decision tree structure layer by layer. The selected features of each layer are based on the segmentation results of the previous layer, ensuring that the final decision tree can effectively classify and predict the functional features of the mini program. In this way, the system can generate a decision tree model with clear hierarchy and excellent classification effect.

[0130] In terms of the technical problems solved, step S402 mainly solves the problem of how to construct an effective decision tree to select features based on feature segmentation points and functional summary features. In the multi-level feature segmentation and selection process, simple feature selection methods may not be able to cope with complex data structures and classification requirements. Through recursive segmentation and Gini impurity calculation, the system can build a decision tree layer by layer to ensure that the selected features at each layer can maximize the classification effect and accuracy of the model.

[0131] In terms of the technical effects achieved, through step S402, the system can generate the final decision tree selection features based on the feature segmentation points and the functional summary features. These selection features can not only accurately reflect the differences between the functional features of the mini-programs, but also effectively guide the development and optimization of the mini-programs. The decision tree model finally generated can accurately classify and predict the functional features of the mini-programs, improving the functional coverage and user experience of the mini-programs.

[0132] For example, continuing with the example of the health monitoring applet above, the system determines the optimal segmentation point of the sensor data acquisition frequency in step S401. In step S402, the system segments and classifies the data acquisition frequencies of different devices and operating systems based on this segmentation point. Then, the system constructs the selection features of the decision tree based on the segmented features and generates the final decision tree model. This decision tree model can accurately classify and predict the health monitoring function features of different devices and operating systems, providing strong support for the development and optimization of the applet.

[0133] In general, through steps S401 and S402, the system can effectively input the mini-program function summary features into the decision tree model, use the Gini impurity calculation to select the optimal feature segmentation point, and generate the final decision tree selection features. This not only improves the accuracy of feature selection and the classification effect of the decision tree model, but also provides strong support for the development and optimization of mini-programs. The mini-programs finally generated can meet the functional requirements of a variety of devices and operating systems, and improve user experience and satisfaction.

[0134] In one embodiment of the integrated development method for mini-programs of the present application, see Figure 5 , and can also include the following:

[0135] Step S501: inputting the decision tree selection feature into the splitting layer of a preset decision tree model, performing a recursive splitting operation according to a preset maximum tree depth, and determining the corresponding decision tree recursive feature;

[0136] Step S502: performing a pruning operation on the decision tree recursive features to determine the decision tree prediction features output by the corresponding preset decision tree model, wherein the decision tree prediction features are used to predict category labels.

[0137] Optionally, in this embodiment, in step S501, the system inputs the decision tree selection features determined in the previous step into the splitting layer of the preset decision tree model. The task of the splitting layer is to construct a decision tree layer by layer through recursive splitting operations based on the input features until the preset maximum tree depth is reached. The specific technical process of implementation is that the system first receives the decision tree selection features, and then starts the recursive splitting operation in the splitting layer. The recursive splitting operation refers to selecting an optimal feature for splitting at each node of the decision tree to minimize the impurity of the data. This process will continue until the decision tree reaches the preset maximum depth, or the data in the node can no longer be effectively split.

[0138] In terms of technical principles, the recursive split operation uses the basic principles of decision trees to build a decision tree layer by layer by continuously selecting features with the lowest Gini impurity for splitting. Each layer of splitting will select a feature that can best distinguish the data category for splitting based on the data on the current node, thereby generating a new child node. The child node will continue to perform the same operation until the preset maximum tree depth is reached. The preset maximum tree depth is to prevent the decision tree from being too complex and causing overfitting problems. By limiting the depth of the tree, it ensures that the decision tree has good generalization ability.

[0139] In terms of the technical problems solved, step S501 mainly solves how to use the selected features to perform effective recursive splitting in the decision tree model, so as to build a decision tree with clear hierarchy and excellent classification effect. Since the input features may contain a large amount of different types of information, simple splitting methods may not be able to effectively utilize these features. Through recursive splitting and Gini impurity calculation, the system can scientifically select the optimal features for splitting, thereby building an efficient decision tree structure.

[0140] In terms of the technical effects achieved, through step S501, the system can effectively input the decision tree selection features into the splitting layer, and use the recursive splitting operation to build a decision tree model with clear hierarchy and reasonable structure. This not only improves the classification effect of the decision tree, but also lays a solid foundation for the subsequent pruning operation. The recursive features of the decision tree finally generated can accurately reflect the relationship between the data and provide strong support for the subsequent category prediction.

[0141] For example, suppose a user wants to develop an intelligent recommendation system, and the system generates decision tree selection features based on the user's browsing history, purchase history, and preference data. In step S501, the system inputs these selection features into the splitting layer of the preset decision tree model. The splitting layer selects the best features for splitting layer by layer through recursive splitting operations. For example, the system may first select the user's browsing history as the splitting feature, and then select the purchase record on the child node for further splitting until the decision tree reaches the preset maximum depth. Through this recursive splitting operation, the system generates a decision tree model that can accurately distinguish the preferences of different users.

[0142] In step S502, the system performs a pruning operation on the recursive features of the decision tree to determine the decision tree prediction features output by the corresponding preset decision tree model. Pruning operation refers to the process of simplifying the decision tree structure by removing some unnecessary nodes after the decision tree is built, thereby improving the generalization ability of the model. The specific technical process is that the system first traverses the constructed decision tree to identify nodes and branches that may cause overfitting. Then, the system gradually removes these nodes and branches according to a preset pruning algorithm (such as cost complexity pruning or minimum error pruning) until the decision tree reaches an optimal equilibrium state.

[0143] In terms of technical principles, pruning reduces the complexity of decision trees to avoid overfitting of the model. During the training process, decision trees may overfit the noise and details in the training data, resulting in poor performance of the model on new data. Through pruning, the system can remove nodes and branches that do not contribute much to the model's prediction effect, thereby improving the model's generalization ability and prediction accuracy.

[0144] In terms of the technical problems to be solved, step S502 mainly solves the problem of how to improve the generalization ability and prediction effect of the model through pruning operations in the decision tree model. In a complex data environment, the decision tree may overfit the training data. Through pruning operations, the system can simplify the decision tree structure and avoid overfitting, thereby improving the performance of the model on new data.

[0145] In terms of the technical effects achieved, through step S502, the system can effectively prune the decision tree recursive features to generate simplified and efficient decision tree prediction features. These prediction features can not only accurately reflect the relationship between data, but also effectively guide the prediction and classification of the model. The decision tree model finally generated has good generalization ability and prediction accuracy, and can run stably in different data environments.

[0146] For example, continuing with the example of the intelligent recommendation system above, the decision tree model generated in step S501 may contain many complex nodes and branches. In step S502, the system prunes these recursive features to remove those nodes and branches that do not contribute much to the prediction effect. For example, if some nodes only contain preference data for a very small number of users, the system may prune these nodes to simplify the decision tree structure. After the pruning operation, the system generates a simplified and efficient decision tree model that can stably perform recommendations and predictions in different user data environments.

[0147] In general, through steps S501 and S502, the system can effectively input the decision tree selection features into the splitting layer, build the decision tree model using recursive splitting operations, and simplify the decision tree structure through pruning operations to generate efficient decision tree prediction features. This not only improves the classification effect and generalization ability of the model, but also provides strong support for the development and optimization of mini programs.

[0148] In one embodiment of the integrated development method for mini-programs of the present application, see Figure 6 , and can also include the following:

[0149] Step S601: input the mini-program function summary features into the feature selection layer of the preset regression tree model, perform feature selection operation according to the preset mean square error algorithm, determine the corresponding regression tree selection features, perform recursive splitting operation according to the regression tree selection features, and determine the corresponding regression tree recursive features;

[0150] Step S602: input the regression tree recursive features into the pruning layer of the preset regression tree model, perform pruning operations on the regression tree recursive features, determine corresponding regression tree prediction features, and input the regression tree prediction features into the feature integration layer of the preset regression tree model, wherein the regression tree prediction features are used to predict continuous values.

[0151] Optionally, in this embodiment, in step S601, the system first inputs the mini-program function summary features into the feature selection layer of the preset regression tree model. The regression tree model is a decision tree model used to predict continuous values. The task of its feature selection layer is to select the features that best explain the data changes through the mean square error algorithm. The specific technical process is that the system receives the input mini-program function summary features and passes them to the feature selection layer. The feature selection layer traverses each feature, calculates its corresponding mean square error (MSE), and selects the features that can minimize the mean square error as the regression tree selection features.

[0152] In terms of technical principles, the mean square error algorithm measures the explanatory power of features by calculating the sum of squared errors of features to predicted values. Specifically, the system splits each feature, calculates the change in mean square error before and after the split, and selects the feature that can minimize the error for splitting. This process ensures that the selected features can effectively reduce the prediction error, thereby improving the prediction accuracy of the model.

[0153] In terms of the technical problems solved, step S601 mainly solves the problem of how to select the features that best explain data changes from the many mini-program function summary features. Since the mini-program function features may contain a large number of different types of features, direct prediction may lead to excessive model complexity or insufficient prediction accuracy. Through the mean square error algorithm, the system can scientifically select the optimal features, thereby ensuring that the model has high prediction accuracy while maintaining low complexity.

[0154] In terms of the technical effects achieved, through step S601, the system can effectively input the mini-program function summary features into the preset regression tree model, and perform feature selection through the mean square error algorithm to generate the corresponding regression tree selection features. Subsequently, the system performs recursive splitting operations based on these selection features, builds the regression tree model layer by layer, and finally generates regression tree recursive features that can accurately predict continuous values. This not only improves the prediction effect of the model, but also lays a solid foundation for subsequent pruning operations.

[0155] For example, suppose a user wants to develop a smart temperature control applet, and the system needs to predict the temperature changes in the future based on environmental data (such as humidity, light intensity, wind speed, etc.). In step S601, the system first inputs these environmental data as the applet function summary features into the feature selection layer of the regression tree model. The feature selection layer selects the features that best explain the temperature changes, such as humidity and light intensity, through the mean square error algorithm. Then, the system performs recursive splitting operations based on these selected features, builds a regression tree model layer by layer, and finally generates a regression tree recursive feature that can accurately predict temperature changes.

[0156] In step S602, the system inputs the recursive features of the regression tree into the pruning layer of the preset regression tree model and performs pruning operations on these recursive features. The pruning operation is to simplify the regression tree structure, avoid overfitting problems, and improve the generalization ability of the model. The specific technical process is that the system first traverses the constructed regression tree to identify nodes and branches that may cause overfitting. Then, the system gradually removes these nodes and branches according to the preset pruning algorithm until the regression tree reaches an optimal balance state.

[0157] In terms of technical principles, pruning reduces the complexity of the regression tree to avoid overfitting of the model. During the training process, the regression tree may overfit the noise and details in the training data, resulting in poor performance of the model on new data. Through pruning, the system can remove nodes and branches that do not contribute much to the model's prediction effect, thereby improving the model's generalization ability and prediction accuracy.

[0158] In terms of the technical problems to be solved, step S602 mainly solves the problem of how to improve the generalization ability and prediction effect of the model through pruning operations in the regression tree model. In a complex data environment, the regression tree model may overfit the training data. Through pruning operations, the system can simplify the regression tree structure and avoid overfitting, thereby improving the performance of the model on new data.

[0159] In terms of the technical effects achieved, through step S602, the system can effectively prune the regression tree recursive features to generate simplified and efficient regression tree prediction features. Then, the system inputs these prediction features into the feature integration layer of the regression tree model to further improve the prediction effect of the model. The regression tree model finally generated has good generalization ability and prediction accuracy, and can run stably in different data environments.

[0160] For example, continuing with the example of the smart temperature control applet above, the regression tree model generated in step S601 may contain many complex nodes and branches. In step S602, the system prunes these recursive features to remove those nodes and branches that do not contribute much to the prediction effect. For example, if some nodes only contain data for very few special environments, the system may prune these nodes to simplify the regression tree structure. After the pruning operation, the system inputs the simplified regression tree prediction features into the feature integration layer to further optimize the model. The final generated regression tree model can accurately predict temperature changes and provide users with intelligent temperature control recommendations.

[0161] In general, through steps S601 and S602, the system can effectively input the mini-program function summary features into the regression tree model, use the mean square error algorithm for feature selection, build a regression tree model, and simplify the regression tree structure through pruning operations to generate efficient regression tree prediction features. This not only improves the prediction effect and generalization ability of the model, but also provides strong support for the development and optimization of mini-programs.

[0162] In one embodiment of the integrated development method for mini-programs of the present application, see Figure 7 , and can also include the following:

[0163] Step S701: performing a weight matching operation according to the decision tree prediction feature and the regression tree prediction feature output by the pruning layer in the preset regression tree model to determine the corresponding decision tree prediction feature weight and regression tree prediction feature weight;

[0164] Step S702: Perform a feature weighted average operation based on the decision tree prediction feature weights and the regression tree prediction feature weights to determine the corresponding mini-program function features.

[0165] Optionally, in this embodiment, in step S701, the system first performs a weight matching operation based on the decision tree prediction features and the regression tree prediction features output by the pruning layer in the preset regression tree model to determine the corresponding decision tree prediction feature weights and regression tree prediction feature weights. The specific technical process includes the following steps: the system receives the prediction features from the decision tree model and the prediction features from the regression tree model. Then, the system performs a weight matching operation on these two sets of features through a series of algorithms (such as a weight optimization algorithm based on cross-validation) to determine the importance of each feature in the final prediction. The core of the weight matching operation is to calculate the contribution of each feature to the prediction of the target variable and assign weights accordingly.

[0166] In terms of technical principles, the weight matching operation utilizes the idea of ​​multi-model integration, and maximizes the accuracy of comprehensive prediction by optimizing the weights of output features of different models. The decision tree and regression tree models have different feature selection and splitting mechanisms. Through weight matching, the advantages of both can be effectively integrated to improve the prediction effect. Specifically, the system calculates the prediction error of each feature and the correlation between features, and assigns weights based on these indicators to ensure that the final weighted prediction result can minimize the overall prediction error.

[0167] In terms of the technical problems solved, step S701 mainly solves how to scientifically allocate the weights of the output features of different models in the process of multi-model integration to improve the accuracy of the final prediction results. Due to the differences in feature selection and prediction mechanisms between decision tree and regression tree models, simply combining the prediction results of the two may result in poor prediction results. Through the weight matching operation, the system can scientifically evaluate and allocate feature weights to ensure the reliability and accuracy of the comprehensive prediction results.

[0168] In terms of the technical effects achieved, through step S701, the system can effectively integrate the prediction features of the decision tree and regression tree models to generate a set of prediction features with scientific weight distribution. This not only improves the prediction accuracy of the comprehensive model, but also lays a solid foundation for the subsequent feature weighted average operation. The decision tree prediction feature weights and regression tree prediction feature weights finally generated can accurately reflect the importance of their respective features in the prediction, thereby optimizing the overall prediction performance.

[0169] For example, suppose a user wants to develop an intelligent health management applet. The system needs to predict health risks based on the user's physiological data (such as heart rate, blood pressure, body temperature, etc.) and lifestyle data (such as diet, exercise, sleep, etc.). In step S701, the system first obtains prediction features from the decision tree model and the regression tree model respectively. Then, the system calculates the weight of each prediction feature through a weight matching operation. For example, the system may find that heart rate has a better prediction effect in the decision tree model, while diet has a better prediction effect in the regression tree model. Through the weight matching operation, the system optimizes the weights of these features to ensure the accuracy of the final prediction results.

[0170] In step S702, the system performs a weighted average operation on the features according to the decision tree prediction feature weights and the regression tree prediction feature weights to determine the corresponding mini program functional features. The specific technical process is that the system first applies the feature weights determined in step S701 to the corresponding predicted features, and then performs a weighted average operation on the weighted features to generate the final mini program functional features.

[0171] In terms of technical principles, the feature weighted average operation generates a comprehensive prediction result by combining the weights of different prediction features. The core of the weighted average operation is to use the weight of each feature to quantify its contribution to the prediction result, thereby generating a more accurate comprehensive prediction result. Specifically, the system multiplies each prediction feature by its corresponding weight, and then adds the weighted feature values ​​to obtain the final mini program function features.

[0172] In terms of the technical problems to be solved, step S702 mainly solves the problem of how to generate a comprehensive prediction result through a weighted average operation in the process of multi-model integration. Since the prediction features of different models have different importance, simply combining these features directly may lead to deviations in the prediction results. Through the feature weighted average operation, the system can scientifically integrate the contributions of different features to generate a more accurate and reliable prediction result.

[0173] In terms of the technical effects achieved, through step S702, the system can effectively perform weighted averaging on the prediction features of the decision tree and regression tree models to generate the final mini-program functional features. This not only improves the prediction effect of the model, but also provides strong support for the development and optimization of the mini-program. The mini-program functional features finally generated can accurately reflect the health risks of users and provide users with scientific health management suggestions.

[0174] For example, continuing with the example of the smart health management applet above, in step S701, the system has determined the weights of features such as heart rate, blood pressure, body temperature, diet, exercise, and sleep. In step S702, the system applies the weights of these features to the corresponding feature values, and performs a weighted average operation to generate a comprehensive health risk prediction feature. This comprehensive feature will be used in the health risk scoring module of the applet to provide users with personalized health management suggestions, such as adjusting diet, increasing exercise time, or improving sleep quality. Through these steps, the system not only improves the accuracy of the prediction, but also provides a more scientific and feasible health management plan.

[0175] In order to be able to carry out integrated development of mini-programs more intelligently, accurately and efficiently, so as to improve development efficiency and mini-program quality, the present application provides an embodiment of an integrated development device for mini-programs for implementing all or part of the contents of the integrated development method for mini-programs, see Figure 8 The integrated development device for mini-programs specifically includes the following contents:

[0176] The function feature collection module 10 is used to receive the mini-program development instruction sent by the user, determine the corresponding device type and operating system environment according to the mini-program development instruction, and determine the corresponding mini-program function summary characteristics according to the device type and operating system environment;

[0177] Function feature prediction module 20, used to input the mini-program function summary features into the feature selection layer of a preset decision tree model, determine the corresponding decision tree selection features according to a preset segmentation rule, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0178] The mini-program construction module 30 is used to input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program functional features based on the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model and the preset model integration algorithm, determine the corresponding mini-program push model based on the mini-program functional features, preset real features and the back propagation algorithm, and perform the mini-program push operation according to the mini-program push model.

[0179] From the above description, it can be seen that the integrated development device for mini-programs provided in the embodiment of the present application can receive mini-program development instructions sent by users, determine the corresponding device type and operating system environment according to the mini-program development instructions, and determine the corresponding mini-program function summary characteristics according to the device type and operating system environment; input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determine the corresponding mini-program function features according to the mini-program function features, preset real features and the back propagation algorithm, and perform the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient mini-program integrated development.

[0180] From the hardware level, in order to be able to carry out integrated development of mini-programs more intelligently, accurately and efficiently, so as to improve development efficiency and mini-program quality, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the integrated development method for mini-programs, and the electronic device specifically includes the following contents:

[0181] Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the integrated development device for mini-programs and related devices such as core business systems, user terminals and related databases; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the integrated development method for mini-programs and the embodiment of the integrated development device for mini-programs in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0182] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0183] In practical applications, part of the integrated development method for mini-programs can be executed on the electronic device side as described above, or all operations can be completed in the client device. The specific selection can be based on the processing power of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0184] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0185] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0186] In one embodiment, the integrated development method function for mini-programs may be integrated into the central processing unit 9100. The central processing unit 9100 may be configured to perform the following control:

[0187] Step S101: receiving a mini-program development instruction sent by a user, determining a corresponding device type and operating system environment according to the mini-program development instruction, and determining a corresponding mini-program function summary feature according to the device type and operating system environment;

[0188] Step S102: Input the mini-program function summary features into the feature selection layer of the preset decision tree model, determine the corresponding decision tree selection features according to the preset segmentation rules, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0189] Step S103: Input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determine the corresponding mini-program function module model according to the mini-program function features, preset real features, and the back propagation algorithm, and perform the mini-program project generation operation according to the mini-program function module model.

[0190] From the above description, it can be seen that the electronic device provided in the embodiment of the present application receives a mini-program development instruction sent by a user, determines the corresponding device type and operating system environment according to the mini-program development instruction, and determines the corresponding mini-program function summary characteristics according to the device type and operating system environment; inputs the decision tree prediction feature into the feature integration layer of the preset regression tree model, determines the corresponding mini-program function characteristics according to the decision tree prediction feature, the regression tree prediction feature output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determines the corresponding mini-program function characteristics according to the mini-program function characteristics, preset real features and the back propagation algorithm, and performs the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient mini-program integrated development.

[0191] In another embodiment, the integrated development device for mini-programs can be configured separately from the central processing unit 9100. For example, the integrated development device for mini-programs can be configured as a chip connected to the central processing unit 9100, and the functions of the integrated development method for mini-programs can be implemented through the control of the central processing unit.

[0192] like Fig. 9 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0193] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device, which receives input and controls the operation of various components of the electronic device 9600.

[0194] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0195] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0196] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0197] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0198] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0199] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0200] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps in the integrated development method for mini-programs in the above embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps in the integrated development method for mini-programs in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0201] Step S101: receiving a mini-program development instruction sent by a user, determining a corresponding device type and operating system environment according to the mini-program development instruction, and determining a corresponding mini-program function summary feature according to the device type and operating system environment;

[0202] Step S102: Input the mini-program function summary features into the feature selection layer of the preset decision tree model, determine the corresponding decision tree selection features according to the preset segmentation rules, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0203] Step S103: Input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determine the corresponding mini-program function module model according to the mini-program function features, preset real features, and the back propagation algorithm, and perform the mini-program project generation operation according to the mini-program function module model.

[0204] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application receives the mini-program development instructions sent by the user, determines the corresponding device type and operating system environment according to the mini-program development instructions, and determines the corresponding mini-program function summary characteristics according to the device type and operating system environment; inputs the decision tree prediction features into the feature integration layer of the preset regression tree model, determines the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determines the corresponding mini-program function features according to the mini-program function features, preset real features and the back propagation algorithm, and performs the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient mini-program integrated development.

[0205] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the integrated development method for mini-programs in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the integrated development method for mini-programs are implemented. For example, the computer program / instruction implements the following steps:

[0206] Step S101: receiving a mini-program development instruction sent by a user, determining a corresponding device type and operating system environment according to the mini-program development instruction, and determining a corresponding mini-program function summary feature according to the device type and operating system environment;

[0207] Step S102: Input the mini-program function summary features into the feature selection layer of the preset decision tree model, determine the corresponding decision tree selection features according to the preset segmentation rules, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features;

[0208] Step S103: Input the decision tree prediction features into the feature integration layer of the preset regression tree model, determine the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determine the corresponding mini-program function module model according to the mini-program function features, preset real features, and the back propagation algorithm, and perform the mini-program project generation operation according to the mini-program function module model.

[0209] From the above description, it can be seen that the computer program product provided in the embodiment of the present application receives the mini-program development instructions sent by the user, determines the corresponding device type and operating system environment according to the mini-program development instructions, and determines the corresponding mini-program function summary characteristics according to the device type and operating system environment; inputs the decision tree prediction features into the feature integration layer of the preset regression tree model, determines the corresponding mini-program function features according to the decision tree prediction features, the regression tree prediction features output by the pruning layer in the preset regression tree model, and the preset model integration algorithm, determines the corresponding mini-program function features according to the mini-program function features, preset real features and the back propagation algorithm, and performs the mini-program project generation operation according to the mini-program function module model, thereby enabling more intelligent, accurate and efficient mini-program integrated development.

[0210] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0211] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0212] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0213] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0214] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. An integrated development method for mini-programs, characterized in that: The method comprises: Receive a mini-program development instruction sent by a user, determine a corresponding device type and operating system environment according to the mini-program development instruction, perform a minority class identification operation on the device type and operating system environment according to a preset minority class identification rule, determine a corresponding nearest neighbor mini-program functional feature according to the minority mini-program functional features obtained after the minority class identification operation and a preset K nearest neighbor algorithm, perform a distance calculation operation on the minority mini-program functional features and the nearest neighbor mini-program functional features to determine a corresponding mini-program functional feature distance, determine a corresponding synthetic mini-program functional feature according to the mini-program functional feature distance and a preset ratio rule, perform a feature addition operation on the mini-program functional features according to the synthetic mini-program functional features, and determine a corresponding mini-program functional summary feature; Input the mini-program function summary features into the feature selection layer of the preset decision tree model, and determine the corresponding feature segmentation point according to the preset Gini impurity; determine the corresponding decision tree selection features according to the feature segmentation points and the mini-program function summary features, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features; The decision tree prediction feature is input into the feature integration layer of the preset regression tree model, and a weight matching operation is performed based on the decision tree prediction feature and the regression tree prediction feature output by the pruning layer in the preset regression tree model to determine the corresponding decision tree prediction feature weight and regression tree prediction feature weight; a feature weighted averaging operation is performed based on the decision tree prediction feature weight and the regression tree prediction feature weight to determine the corresponding mini-program functional feature, the corresponding mini-program functional module model is determined based on the mini-program functional feature, the preset real feature and the back propagation algorithm, and the mini-program project generation operation is performed based on the mini-program functional module model.

2. The integrated development method for mini-programs according to claim 1, characterized in that: The step of inputting the decision tree selection feature into a splitting layer of a preset decision tree model to perform a recursive splitting operation, determining a corresponding decision tree recursive feature, and determining a decision tree prediction feature output by a corresponding preset decision tree model according to the decision tree recursive feature, comprises: Inputting the decision tree selection features into the splitting layer of a preset decision tree model, performing a recursive splitting operation according to a preset maximum tree depth, and determining corresponding decision tree recursive features; The decision tree recursive features are pruned to determine the decision tree prediction features output by the corresponding preset decision tree model, wherein the decision tree prediction features are used to predict category labels.

3. The integrated development method for mini-programs according to claim 1, characterized in that: Before inputting the decision tree prediction features into the feature integration layer of the preset regression tree model, the method includes: Input the mini-program function summary features into the feature selection layer of a preset regression tree model, perform feature selection operations according to a preset mean square error algorithm, determine corresponding regression tree selection features, perform recursive splitting operations according to the regression tree selection features, and determine corresponding regression tree recursive features; The regression tree recursive features are input into the pruning layer of the preset regression tree model, the regression tree recursive features are pruned, the corresponding regression tree prediction features are determined, and the regression tree prediction features are input into the feature integration layer of the preset regression tree model, wherein the regression tree prediction features are used to predict continuous values.

4. An integrated development device for mini-programs, characterized in that: The device comprises: A function feature acquisition module is used to receive a mini-program development instruction sent by a user, determine a corresponding device type and operating system environment according to the mini-program development instruction, perform a minority class identification operation on the device type and operating system environment according to a preset minority class identification rule, determine a corresponding nearest neighbor mini-program function feature according to the minority mini-program function feature obtained after the minority class identification operation and a preset K nearest neighbor algorithm, perform a distance calculation operation on the minority mini-program function feature and the nearest neighbor mini-program function feature to determine a corresponding mini-program function feature distance, determine a corresponding synthetic mini-program function feature according to the mini-program function feature distance and a preset ratio rule, perform a feature addition operation on the mini-program function feature according to the synthetic mini-program function feature, and determine a corresponding mini-program function summary feature; A function feature prediction module is used to input the function summary features of the mini program into the feature selection layer of a preset decision tree model, and determine the corresponding feature segmentation point according to the preset Gini impurity; determine the corresponding decision tree selection features according to the feature segmentation point and the function summary features of the mini program, input the decision tree selection features into the splitting layer of the preset decision tree model for recursive splitting operation, determine the corresponding decision tree recursive features, and determine the corresponding decision tree prediction features output by the preset decision tree model according to the decision tree recursive features; The mini program construction module is used to input the decision tree prediction features into the feature integration layer of the preset regression tree model, perform weight matching operations based on the decision tree prediction features and the regression tree prediction features output by the pruning layer in the preset regression tree model, and determine the corresponding decision tree prediction feature weights and regression tree prediction feature weights; perform feature weighted averaging operations based on the decision tree prediction feature weights and the regression tree prediction feature weights to determine the corresponding mini program functional features, determine the corresponding mini program functional module model based on the mini program functional features, preset real features and the back propagation algorithm, and execute the mini program project generation operation based on the mini program functional module model.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the integrated development method for mini-programs described in any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the integrated development method for mini-programs described in any one of claims 1 to 3 are implemented.

Images