A method for automatically generating corresponding models based on data analysis

Abstract

The invention discloses a method for automatically generating a corresponding model based on data analysis, which belongs to the field of artificial intelligence technology and includes a system composition, wherein the system composition includes a data capture module and a model processing module, wherein the data capture module includes a capture request, a capture data processing track and a monitoring data return; the model processing module includes model generation and model fine-tuning; the capture request is based on existing platform data, and the data processing frequently operated by the user is judged through the capture data processing track and the monitoring data return, and the model processing module self-analyzes and processes, automatically generates a corresponding model business, and mounts it to a business model library, and the user directly operates the corresponding model according to the needs. The application can automatically learn the user's business operation habits, and generate corresponding model functions according to the operations, thereby reducing the user's operation costs and ensuring data security; the business function automatically generates the model, and the data is not exposed in combination with the model fine-tuning.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and in particular relates to a method for automatically generating a corresponding model based on data analysis. Background Art

[0002] Nowadays, when users handle some common businesses, they often frequently operate complex businesses. The main reasons include: 1. Users operate complex businesses by themselves, which is time-consuming and labor-intensive each time; 2. Use other software to complete the current work, but this will lead to data leakage and make it difficult to ensure data security; 3. Contact the development company to customize business needs, but this will increase costs. There are many paths for a method to be executed, and it is impossible to know where the method is executed.

[0003] Existing technologies in terms of performance and cost: Although existing middleware generally has good performance, it may not achieve optimal results in certain scenarios. When the project business performance is not met, the enterprise may need to purchase additional support services or conduct secondary development to improve performance, which will incur costs.

[0004] Existing technologies in terms of customizability: They are usually standardized products. Although they provide certain configuration options, they may not be flexible enough in the face of highly customized requirements. When the project business is not met, customized business functions may not be met.

[0005] The security of existing technologies depends on the implementation of the vendor and the operation and maintenance level of the enterprise. Although most well-known middlewares have good security mechanisms, there may still be undiscovered vulnerabilities. Because there are references to other plug-ins or components, there are unknown vulnerabilities, which may cause security issues when using existing middleware.

[0006] Therefore, it is crucial to develop a method to automatically generate corresponding models based on data analysis. Summary of the invention

[0007] The purpose of the present invention is to provide a method for automatically generating a corresponding model based on data analysis to solve the problems raised by the above background technology.

[0008] A method for automatically generating a corresponding model based on data analysis of the present invention includes a system composition, wherein the system composition includes a data capture module and a model processing module, wherein the data capture module includes a capture request, a capture data processing track, and a monitoring data return; the model processing module includes model generation and model fine-tuning; the capture request is based on existing platform data, and the data processing frequently operated by the user is judged through the capture data processing track and the monitoring data return, and the model processing module self-analyzes and processes, automatically generates a corresponding model business, and mounts it to a business model library, and the user directly operates the corresponding model according to the needs.

[0009] Furthermore, the capture request intercepts the user request through the springboot interceptor and stores it in mongodb, and generates a capture line list corresponding to the interval according to the user request.

[0010] Furthermore, the captured data processing trajectory uses WxStackIntercept to intercept the data and monitor the processing of business data by the business implementation layer, which is used to solve the confirmation of a method execution path, and is stored in the corresponding business trajectory table in the form of an observation expression; WxStackIntercept intercepts and processes data according to the upper layer method, searches upward layer by layer until the top-level class is found, and generates the corresponding specific business trajectory line according to the business class reflection; and the result falls into mongodb.

[0011] Furthermore, the WxStackIntercept interception step - the step of interface level request is:

[0012] S1. This class first implements HandlerInterceptor, and rewrites the preHandle method to implement the pre-interception of each request (hereinafter referred to as WXHIPH); rewrites the afterCompletion method to implement the post-return interception of each request (hereinafter referred to as WXHIAC);

[0013] S2. Add an interception mark to the service interface address field that needs to be monitored;

[0014] S3. The interface address obtained from WXHIPH is determined by contain whether it belongs to the model class. If it does, the request is marked as a record and entered into mongodb for follow-up until the request is completed.

[0015] S4. When receiving the WXHIAC callback, it indicates that the current task has been completed, and all data tracks from the beginning to the end are stored in the database.

[0016] Furthermore, the WxStackIntercept interception step - the method level request step is:

[0017] S1. Create an aspect class through @Aspect and add it to the Spring container through the @Component annotation;

[0018] S2. Through @Before and @After, according to the package scope corresponding to the model, this aspect will record the entry and exit method level events of all methods matching the expression;

[0019] S3. The data intercepted at the method level enters MongoDB as a follow-up until the request is completed.

[0020] Furthermore, the logical steps of the WxStackIntercept are:

[0021] S1. The home page automatically combines the corresponding business process records into a set of tracks based on the start time and end time of the business interface level of the same business, and labels them with RT;

[0022] S2. Identify the corresponding class description and method description of the current class path at the corresponding method level according to reflection;

[0023] S3, after multiple collections of the same business, automatically combine them into a model;

[0024] S4. The parameters of the corresponding trajectory are also adjusted.

[0025] Furthermore, the automatic combination into a model is replaced by manually grouping the corresponding groups into a custom model.

[0026] Furthermore, the monitoring data returns the track data of the captured data processing track and returns it to MongoDB.

[0027] Furthermore, model generation: according to the capture line list generated by the capture data module each time, a corresponding model is generated through data analysis and mounted into the corresponding model module.

[0028] Furthermore, the model is fine-tuned to meet the user's requirements through the model learned by the business itself combined with the user's customized modification of model parameters. The customized modification of model parameters includes model processing time and model business classification, which is more in line with the user's system. The user specifies the model trigger time and trigger parameters to achieve the user's own customized behavior.

[0029] The beneficial effects of the technical solution of the present invention are:

[0030] 1. This application determines the data processing that users frequently operate based on the existing platform data, automatically generates the corresponding model business, and mounts it into the business system. Subsequent users can directly operate the corresponding model as needed to reduce the user's workload.

[0031] 2. This application can automatically learn the user's business operation habits and generate corresponding model functions based on the operations, reducing the user's operating costs and ensuring data security; it can realize the automatic generation of models by business functions and combine model fine-tuning to ensure that data is not leaked.

[0032] 3. This application uses the self-developed WxStackIntercept as the data interception monitoring code to process the business data at the business implementation layer. In terms of performance, it can be deeply optimized according to specific business scenarios, solve performance bottleneck problems in a targeted manner, and improve the system's response speed and processing capabilities. In terms of customizability, it can be completely customized according to the needs of the enterprise, including functions, interfaces, data formats, etc., to meet the personalized business needs of the enterprise. In terms of security, security can be taken as the primary consideration in the design and implementation stage, and the security of data transmission and storage can be ensured through encryption, access control, audit logs and other means. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a structural diagram of a method for automatically generating a corresponding model based on data analysis as described in the present invention. DETAILED DESCRIPTION

[0034] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0036] The following is a detailed description of a method for automatically generating a corresponding model based on data analysis provided by the present invention in conjunction with the accompanying drawings.

[0037] Refer to the attached Figure 1 , which shows a method for automatically generating corresponding models based on data analysis provided by an embodiment of the present invention, including system components, the system components including a data capture module and a model processing module, the data capture module including a capture request, a capture data processing track and a monitoring data return; the model processing module includes model generation and model fine-tuning; the capture request is based on existing platform data, and the data processing frequently operated by the user is judged by the capture data processing track and the monitoring data return, and the model processing module self-analyzes and processes, automatically generates a corresponding model business, and mounts it to the business model library, and the user directly operates the corresponding model according to the needs.

[0038] The capture request intercepts the user request through the springboot interceptor and stores it in mongodb, and generates a capture line list corresponding to the interval according to the user request.

[0039] The captured data processing track is used to intercept the business data of the business implementation layer through WxStackIntercept, which is used to confirm the execution path of a method, and is stored in the corresponding business track table by observing the form of the expression: WxStackIntercept demo.moxingyi pt 'params[0]<0'; WxStackIntercept intercepts the data processing method of the upper layer, searches upward layer by layer, and searches the top class, and generates the corresponding specific business track line according to the business class reflection; and the result falls into mongodb.

[0040] The WxStackIntercept interception steps - the steps of interface level request are:

[0041] S1. This class first implements HandlerInterceptor, and rewrites the preHandle method to implement the pre-interception of each request (hereinafter referred to as WXHIPH); rewrites the afterCompletion method to implement the post-return interception of each request (hereinafter referred to as WXHIAC);

[0042] S2. Add the interception mark registry.addInterceptor(customInterceptor).excludePathPatterns(" / moxingyi").addPathPatterns(" / *") in the business interface address field that needs to be monitored and processed;

[0043] S3. Get the interface address from WXHIPH and determine whether it belongs to the model class according to contain. If it does, mark the request to form a record [{"mxKey":"moxingyi","url":" / xxx / xxx","iType":"in","createTime":"1728372359","params":{"param1","zhangsan":"param2","2019":"param3","birtd":"1"},"uid":"123122222","sn":"1 2311222111", "parNoteSn": "12311222111", "taskSn": "1111"] enter into mongodb (mxKey means: business key, createTime means: occurrence time, params means: request parameters in the interface, uid means: current request user flag, sn means: current request serial number, parNoteSn means: previous node request serial number, iType means: current interception is interface interception, url means: current user request address, taskSn means: task unique code) as a follow-up until the request is completed;

[0044] S4. When receiving the WXHIAC callback, it indicates that the current task has been completed, and all data tracks from the beginning to the end are stored in the database.

[0045] The WxStackIntercept interception steps - the steps of method level request are:

[0046] S1. Create an aspect class through @Aspect and add it to the Spring container through the @Component annotation;

[0047] S2. Through @Before

[0048] ("execution(* com.weixiaoinfo.moxing..*.*(..))"),@After("execution(*com.weixiaoinfo.moxing..*.*(..))") According to the package scope corresponding to the model, this aspect will record the entry and exit method-level events of all methods matching the expression (recording method parameters, return values, and exception information);

[0049] S3. Data intercepted at the method level

[0050] [{"mxKey":"moxingyi","iType":"fu","funName":"makeDate","funClass":"com.weixiaoinfo.fun.makeDate","createTime":"1728 372359","params":{"param1","zhangsan":"param2","2019":"param3","birtd":"1"},"uid":"123122222","sn":"12311222111","pa rNoteSn":"12311222111","taskSn":"1111"] enters mongodb (mxKey means: business key, createTime means: occurrence time, params means: request parameters in the interface, uid means: current request user flag, sn means: current request serial number, parNoteSn means: previous node request serial number, iType means: current interception is method interception, funName means: current method name, funClass means: current class path, taskSn means: task unique code) as a follow-up until the request is completed.

[0051] The logical steps of the WxStackIntercept are:

[0052] S1. The home page automatically combines the corresponding business process records into a set of tracks based on the start time and end time of the business interface level of the same business, and labels them with RT;

[0053] S2. The current class path at the corresponding method level can be used to identify the corresponding class description and method description based on reflection;

[0054] S3, after multiple collections of the same business, automatically combine them into a model;

[0055] S4. The parameters of the corresponding trajectory can also be adjusted.

[0056] The automatic combination into a model is replaced by manually grouping the corresponding groups into a custom model.

[0057] The monitoring data returns the track data returned by capturing the data processing track and falls into MongoDB.

[0058] Model generation: Based on the capture line list generated by the capture data module each time, the corresponding model is generated through data analysis (user input habits, business processing time habits, business classification habits) and mounted into the corresponding model module.

[0059] The model fine-tuning meets the user's requirements by combining the model learned by the business itself with the user's customized modification of model parameters. The customized modification of model parameters includes model processing time, model business classification, etc., which is more in line with the user's system. The user can also specify the model trigger time and trigger parameters to achieve the user's own customized behavior.

[0060] Because the generated model is within the current system business, it will not cause data leakage, thus ensuring data security.

[0061] Case 1: Manually group the corresponding sets into a custom model.

[0062] Scenario: A user queries whether a certain person can receive a pension. (The platform includes: querying whether a person is dead based on his / her ID number, querying the payment period, and querying whether the two identity elements are correct)

[0063] Normal business logic:

[0064] Platform staff need to first enter the "check whether the two identity elements are correct" function to determine whether the name and ID card are correct.

[0065] After it is correct, execute the "Query Payment Years Function" to determine whether the current person's payment years meet the requirements.

[0066] After it is correct, execute the "Query whether the person is dead based on the ID number" function to determine whether the current person is alive.

[0067] If all the above conditions are met, you can return and claim it.

[0068] If there is no advance situation of "using self-developed WxStackIntercept to intercept and monitor data and process business data at the business implementation layer":

[0069] Users need to perform four steps, which increases their workload; or they need to develop a corresponding "pension query function", which will increase capital investment and development cycle.

[0070] If there is an advance situation of "using self-developed WxStackIntercept to intercept and monitor data and process business data at the business implementation layer":

[0071] The system will automatically capture all user operations and, based on the user's execution frequency, automatically determine that this is a function that can be merged. Then, the corresponding "pension query function" model will be automatically generated in the system model library. When the user subsequently enters the corresponding model to input the query content, the system will automatically execute the above business logic and return the final result, which can greatly reduce the user's operating costs and learning costs.

[0072] This application uses the self-developed WxStackIntercept as the data interception monitoring code to process the business data at the business implementation layer. In terms of performance, it can be deeply optimized according to specific business scenarios, solve performance bottleneck problems in a targeted manner, and improve the system's response speed and processing capabilities. In terms of customizability, it can be completely customized according to the needs of the enterprise, including functions, interfaces, data formats, etc., to meet the personalized business needs of the enterprise. In terms of security, security can be taken as the primary consideration in the design and implementation stage, and the security of data transmission and storage can be ensured through encryption, access control, audit logs and other means.

[0073] This application determines the data processing that users frequently perform based on the existing platform data, automatically generates the corresponding model business, and mounts it into the business system. Subsequent users can directly operate the corresponding model as needed to reduce the user's workload.

[0074] This application can automatically learn the user's business operation habits and generate corresponding model functions based on the operations, reducing the user's operating costs and ensuring data security; it can realize automatic generation of models based on business functions and combine model fine-tuning to ensure that data is not leaked.

[0075] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0076] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for automatically generating a corresponding model based on data analysis, characterized in that: The system comprises a data capture module and a model processing module. The data capture module comprises a capture request, a capture data processing track and a monitoring data return; the model processing module comprises a model generation and a model fine-tuning; the capture request is based on the existing platform data, and the data processing that the user frequently operates is judged by the capture data processing track and the monitoring data return. The model processing module self-analyzes and processes, and automatically generates a corresponding model business, which is mounted in the business model library. The user directly operates the corresponding model according to the needs. The capture data processing track is intercepted by the WxStackIntercept to intercept the data and the monitoring code to process the business data of the business implementation layer, which is used to solve the confirmation of a method execution path, and is stored in the corresponding business track table in the form of an observation expression; the WxStackIntercept intercepts the method of processing data according to the upper layer, searches upward layer by layer, and searches until the top class is found, and generates a corresponding specific business track line according to the business class reflection; and the result falls into mongodb. The WxStackIntercept interception step-the step of interface level request is as follows: S1. This class first implements HandlerInterceptor, and rewrites the preHandle method to implement the pre-interception of each request (hereinafter referred to as WXHIPH); rewrites the afterCompletion method to implement the post-return interception of each request (hereinafter referred to as WXHIAC); S2. Add an interception mark to the service interface address field that needs to be monitored; S3. The interface address obtained from WXHIPH is determined by contain whether it belongs to the model class. If it does, the request is marked as a record and entered into mongodb for follow-up until the request is completed. S4. When receiving the WXHIAC callback, it indicates that the current task has been completed, and all data tracks from the beginning to the end are stored in the database.

2. The method for automatically generating a corresponding model based on data analysis according to claim 1, characterized in that: The capture request intercepts the user request through the springboot interceptor and stores it in mongodb, and generates a capture line list corresponding to the interval according to the user request.

3. The method for automatically generating a corresponding model based on data analysis according to claim 1, characterized in that: The WxStackIntercept interception steps - the steps of method level request are: S1. Create an aspect class through @Aspect and add it to the Spring container through the @Component annotation; S2. Through @Before and @After, according to the package scope corresponding to the model, this aspect will record the entry and exit method level events of all methods matching the expression; S3. The data intercepted at the method level enters MongoDB as a follow-up until the request is completed.

4. The method for automatically generating a corresponding model based on data analysis according to claim 3, characterized in that: The logical steps of the WxStackIntercept are: S1. The home page automatically combines the corresponding business process records into a set of tracks according to the start time and end time of the business interface level of the same business, and labels them with RT; S2. Identify the corresponding class description and method description of the current class path at the corresponding method level according to reflection; S3, after multiple collections of the same business, automatically combine them into a model; S4. The parameters of the corresponding trajectory are also adjusted.

5. The method for automatically generating a corresponding model based on data analysis according to claim 4, characterized in that: The automatic combination into a model is replaced by manually grouping the corresponding groups into a custom model.

6. The method for automatically generating a corresponding model based on data analysis according to claim 1, characterized in that: The monitoring data returns the track data returned by capturing the data processing track and falls into MongoDB.

7. The method for automatically generating a corresponding model based on data analysis according to claim 1, characterized in that: Model generation: Based on the capture line list generated by the capture data module each time, the corresponding model is generated through data analysis and mounted into the corresponding model module.

8. The method for automatically generating a corresponding model based on data analysis according to claim 1, characterized in that: The model fine-tuning is to meet the user's requirements by combining the model learned by the business itself with the user's customized modification of model parameters. The customized modification of model parameters includes model processing time, model business classification, user-specified model trigger time, and trigger parameters.

Images