Intelligent analysis method, system and medium for supervision problem rectification based on large model

Abstract

This application provides a method, system and medium for intelligent analysis of supervision problem rectification based on a large model. The method includes: obtaining historical supervision data and classifying it, inserting class labels after classification and obtaining class label data sets, obtaining current supervision data and extracting features, inputting current feature data into class label data sets for matching and identification, and processing to obtain target historical supervision data, obtaining a rectification strategy prediction model based on target historical supervision data processing, inputting current feature data into the rectification strategy prediction model for processing, obtaining recommended rectification strategy data and pushing it to the client, and evaluating the supervision subject based on the historical supervision data of the supervision subject, screening out target supervision issues and marking them based on the evaluation of the impact of different supervision issues on the company's operations; thereby realizing intelligent analysis of supervision problem rectification and improving supervision efficiency and company management level.

Description

Technical Field

[0001] The present application relates to the technical field of big data and supervision problem rectification, and specifically, to an intelligent analysis method, system and medium for supervision problem rectification based on a big model. Background Art

[0002] As the scale of enterprises expands and their businesses become more complex, supervision and internal control have become important links to ensure compliance and improve management efficiency. However, traditional supervision problem rectification methods often rely on manual summary, analysis and tracking, which requires a lot of manpower and is inefficient. In addition, the supervision and rectification efforts of companies at different levels are uneven, making it difficult to achieve "similar cases with the same judgment", resulting in units with low rectification efforts being prone to inadequate rectification of problems. In addition, there is a lack of intelligent reminders for whether the same supervisory subject has historical supervision problems, which is not conducive to considering the handling of the supervisory subject.

[0003] In view of the above problems, effective technical solutions are urgently needed. Summary of the invention

[0004] The purpose of this application is to provide a large-model-based intelligent analysis method, system and medium for supervision problem rectification. First, a rectification strategy prediction model is trained based on historical cases of similar supervision problems that have been resolved in the past. In the supervision problem rectification link, the rectification strategy prediction model is used to intelligently recommend supervision rectification methods, and the supervision subject is evaluated based on the historical supervision data of the supervision subject, so as to assist the company in making decisions when handling it. Finally, based on the evaluation of the impact of different supervision issues on the company's operations, target supervision issues with greater impact on the company's operations are selected. This will achieve intelligent analysis of supervision problem rectification and improve supervision efficiency and company management level.

[0005] This application also provides an intelligent analysis method for supervision problem rectification based on a large model, including the following steps:

[0006] Obtain historical supervision data and classify it, and insert class labels into the classified historical supervision data to obtain a class label data set;

[0007] Obtaining current supervision data and performing feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data;

[0008] A rectification strategy prediction model is obtained according to the target historical supervision data processing, the current feature data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client.

[0009] Optionally, in the large model-based intelligent analysis method for supervision problem rectification described in the present application, the acquisition and classification of historical supervision data, and the insertion of class labels into the classified historical supervision data to obtain a class label data set include:

[0010] Obtain historical supervision data and its corresponding company-level data, identify features of historical supervision data, and obtain feature data, including supervision subject data and supervision problem data;

[0011] The historical supervision data is classified according to the supervision problem data to obtain historical supervision data clusters of different categories, and class labels are inserted into the historical supervision data clusters to obtain a class label data set.

[0012] Optionally, in the large model-based supervision problem rectification intelligent analysis method described in the present application, the current supervision data is obtained and feature extraction is performed to obtain the current feature data, the current feature data is input into the class label data set for matching and identification, and the target historical supervision data is obtained by processing, including:

[0013] Obtain current supervision data and its corresponding current company-level data, perform feature extraction on the current supervision data, and obtain current feature data, including current supervision subject data and current supervision problem data;

[0014] The current supervision problem data is input into the class label data set for matching and identification, and the historical supervision data cluster corresponding to the class label that meets the preset matching and identification result is used as the target historical supervision data cluster;

[0015] The current company-level data is compared with the company-level data corresponding to all historical supervision data in the target historical supervision data cluster in sequence, and the historical supervision data corresponding to the company-level data that meets the preset comparison requirements is used as the target historical supervision data.

[0016] Optionally, in the intelligent analysis method for rectification of supervision problems based on a large model described in the present application, the rectification strategy prediction model is obtained according to the target historical supervision data processing, the current feature data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client, including:

[0017] Extract problem rectification strategy data from the target historical supervision data, generate a supervision problem sample set from the supervision problem data set corresponding to all the target historical supervision data, and generate a problem rectification strategy sample set from the problem rectification strategy data set corresponding to all the target historical supervision data;

[0018] Using a preset machine learning algorithm to train the supervised problem sample set and the problem rectification strategy sample set to obtain a rectification strategy prediction model;

[0019] The current supervision problem data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client.

[0020] Optionally, the intelligent analysis method for supervision problem rectification based on a large model described in the present application further includes:

[0021] Matching and identifying the current supervision subject data with all the supervision subject data in sequence, taking the historical supervision data corresponding to the supervision subject data that meets the preset matching and identification requirements as the subject supervision data, and taking the number of supervision subjects that meet the preset matching and identification requirements as the supervision frequency data;

[0022] The supervision problem data corresponding to all the subject supervision data are matched and identified with the current supervision problem data in sequence, and the number of supervision problem data that meets the preset matching and identification requirements is used as the number of similar case supervision times data.

[0023] Optionally, the intelligent analysis method for supervision problem rectification based on a large model described in the present application further includes:

[0024] Extracting problem severity data and problem impact data based on the subject supervision data;

[0025] Calculate based on the problem severity data, problem impact data, supervision frequency data, and similar case supervision frequency data to obtain the supervision subject evaluation index;

[0026] The calculation formula of the supervisory subject evaluation index is:

[0027] ;

[0028] in, To monitor the subject evaluation index, For problem severity data, For the problem impact data, is the number of supervision data, is the number of similar case supervision data, , , , and is the preset characteristic coefficient;

[0029] The supervisory subject evaluation index is compared with a preset supervisory subject evaluation index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the supervisory subject is marked and a reminder message is sent.

[0030] Optionally, the intelligent analysis method for supervision problem rectification based on a large model described in the present application further includes:

[0031] Obtain the number of supervised problem data in the historical supervised data cluster, compare the number of supervised problem data with the number of supervised problem data threshold, and use the historical supervised data cluster corresponding to the number of supervised problem data that meets the preset threshold comparison requirement as the target supervised data cluster;

[0032] Using the supervised problem data corresponding to the target supervised data cluster as target supervised problem data;

[0033] Obtain data on the number of problem supervisors, problem frequency data, problem impact scope data, and problem controversy degree data of the target supervision problem data;

[0034] Calculate the company's operational impact index based on the problem supervision data, problem frequency data, problem impact scope data and problem controversy level data;

[0035] The calculation formula of the company's operational impact index is:

[0036] ;

[0037] in, The company's operational impact index. The number of supervisors for the problem is is the problem frequency data, For the problem impact range data, Data on the controversy level of the issue, , , and is the preset characteristic coefficient;

[0038] The company operation impact index is compared with a preset company operation impact index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the target supervision problem data is marked.

[0039] In a second aspect, the present application provides a supervision problem rectification intelligent analysis system based on a large model, the system comprising: a memory and a processor, the memory storing a program of a supervision problem rectification intelligent analysis method based on a large model, and the program of the supervision problem rectification intelligent analysis method based on a large model is executed by the processor to implement the following steps:

[0040] Obtain historical supervision data and classify it, and insert class labels into the classified historical supervision data to obtain a class label data set;

[0041] Obtaining current supervision data and performing feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data;

[0042] A rectification strategy prediction model is obtained according to the target historical supervision data processing, the current feature data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client.

[0043] Optionally, in the large model-based supervision problem rectification intelligent analysis system described in the present application, the acquisition and classification of historical supervision data, and the insertion of class labels into the classified historical supervision data to obtain a class label data set include:

[0044] Obtain historical supervision data and its corresponding company-level data, identify features of historical supervision data, and obtain feature data, including supervision subject data and supervision problem data;

[0045] The historical supervision data is classified according to the supervision problem data to obtain historical supervision data clusters of different categories, and class labels are inserted into the historical supervision data clusters to obtain a class label data set.

[0046] In the third aspect, the present application also provides a computer-readable storage medium, which stores a program for an intelligent analysis method for rectification of supervisory problems based on a large model. When the program for the intelligent analysis method for rectification of supervisory problems based on a large model is executed by a processor, the steps of the intelligent analysis method for rectification of supervisory problems based on a large model as described in any one of the above items are implemented.

[0047] As can be seen from the above, the intelligent analysis method, system and medium for the rectification of supervision problems based on a large model provided by this application first train a rectification strategy prediction model based on historical cases of similar supervision problems that have been completed in the past. In the rectification of supervision problems, the rectification strategy prediction model is used to intelligently recommend supervision rectification methods, and the supervision subject is evaluated based on the historical supervision data of the supervision subject, so as to assist the company in making decisions when handling it. Finally, the target supervision problems with greater impact on the company's operations are screened out based on the evaluation of the impact of different supervision problems on the company's operations. This realizes the intelligent analysis of the rectification of supervision problems and improves the supervision efficiency and company management level.

[0048] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the embodiments of the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A flowchart of a large-model-based intelligent analysis method for supervision problem rectification provided in an embodiment of the present application;

[0051] Figure 2 A flowchart of obtaining a class label data set for the intelligent analysis method for supervision problem rectification based on a large model provided in an embodiment of the present application;

[0052] Figure 3 A flowchart of obtaining target historical supervision data for the intelligent analysis method for supervision problem rectification based on a large model provided in an embodiment of the present application;

[0053] Figure 4 A flow chart of obtaining recommended rectification strategy data for the large-model-based intelligent analysis method for rectification of supervisory issues provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The technical solutions 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 only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0055] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0056] Please refer to Figure 1 , Figure 11 is a flow chart of a large-model-based intelligent analysis method for supervision problem rectification in some embodiments of the present application. The large-model-based intelligent analysis method for supervision problem rectification is used in terminal devices, such as computers, mobile phone terminals, etc. The large-model-based intelligent analysis method for supervision problem rectification includes the following steps:

[0057] S11, obtaining historical supervision data and classifying it, and inserting class labels into the classified historical supervision data to obtain a class label data set;

[0058] S12, obtaining current supervision data and performing feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data;

[0059] S13, obtaining a rectification strategy prediction model based on the target historical supervision data, inputting the current feature data into the rectification strategy prediction model for processing, obtaining recommended rectification strategy data and pushing it to the client.

[0060] It should be noted that first, a rectification strategy prediction model is trained based on historical cases of similar supervision issues that have been resolved in the past. In the rectification phase of supervision issues, the rectification strategy prediction model is used to intelligently recommend supervision rectification methods, and the supervision subject is evaluated based on the historical supervision data of the supervision subject, so as to assist the company in making decisions when handling it. Finally, the target supervision issues with greater impact on the company's operations are screened out based on the evaluation of the impact of different supervision issues on the company's operations. This will achieve intelligent analysis of the rectification of supervision issues and improve supervision efficiency and company management level.

[0061] Please refer to Figure 2 , Figure 2 The flowchart of obtaining a class label data set of a large model-based supervision problem rectification intelligent analysis method in some embodiments of the present application. According to an embodiment of the present invention, the acquisition of historical supervision data and classification, and the insertion of class labels into the classified historical supervision data to obtain a class label data set include:

[0062] S21. Obtain historical supervision data and its corresponding company-level data, perform feature recognition on the historical supervision data, and obtain feature data, including supervision subject data and supervision problem data;

[0063] S22. Classify the historical supervision data according to the supervision problem data to obtain historical supervision data clusters of different categories, and insert class labels into the historical supervision data clusters to obtain class label data sets.

[0064] It should be noted that historical supervision data at different levels are classified and inserted into class labels according to the supervision problem to facilitate subsequent retrieval.

[0065] Please refer to Figure 3 , Figure 3 The flowchart of the intelligent analysis method for supervision problem rectification based on a large model in some embodiments of the present application is to obtain the target historical supervision data. According to the embodiment of the present invention, the current supervision data is obtained and feature extraction is performed to obtain the current feature data, the current feature data is input into the class label data set for matching and identification, and the target historical supervision data is obtained by processing, including:

[0066] S31, obtaining current supervision data and its corresponding current company-level data, extracting features from the current supervision data, and obtaining current feature data, including current supervision subject data and current supervision problem data;

[0067] S32, inputting the current supervision problem data into the class label data set for matching and identification, and taking the historical supervision data cluster corresponding to the class label that meets the preset matching and identification result as the target historical supervision data cluster;

[0068] S33, sequentially comparing the current company-level data with the company-level data corresponding to all historical supervision data in the target historical supervision data cluster, and taking the historical supervision data corresponding to the company-level data that meets the preset comparison requirements as the target historical supervision data.

[0069] It should be noted that the current supervision problem data is extracted based on the current supervision data and input into the class label data set for matching and identification to obtain historical supervision data of the same type of supervision problem, and locate the target historical supervision data based on the company level where the current supervision data is located.

[0070] Please refer to Figure 4 , Figure 4 The flowchart of obtaining recommended rectification strategy data of the intelligent analysis method for rectification of supervision problems based on a large model in some embodiments of the present application. According to an embodiment of the present invention, the rectification strategy prediction model is obtained according to the target historical supervision data processing, the current feature data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client, including:

[0071] S41, extracting problem rectification strategy data from the target historical supervision data, generating a supervision problem sample set from the supervision problem data set corresponding to all the target historical supervision data, and generating a problem rectification strategy sample set from the problem rectification strategy data set corresponding to all the target historical supervision data;

[0072] S42, using a preset machine learning algorithm to train the supervision problem sample set and the problem rectification strategy sample set to obtain a rectification strategy prediction model;

[0073] S43: Input the current supervision problem data into the rectification strategy prediction model for processing, obtain recommended rectification strategy data and push it to the client.

[0074] It should be noted that the supervision problem data set corresponding to all target historical supervision data is used to generate a supervision problem sample set, and the problem rectification strategy data set corresponding to all target historical supervision data is used to generate a problem rectification strategy sample set. Training is performed based on the supervision problem sample set and the problem rectification strategy sample set to obtain a rectification strategy prediction model, and the current supervision problem data is input into the rectification strategy prediction model for processing to obtain the recommended rectification strategy and push it to the client.

[0075] According to an embodiment of the present invention, it also includes:

[0076] Matching and identifying the current supervision subject data with all the supervision subject data in sequence, taking the historical supervision data corresponding to the supervision subject data that meets the preset matching and identification requirements as the subject supervision data, and taking the number of supervision subjects that meet the preset matching and identification requirements as the supervision frequency data;

[0077] The supervision problem data corresponding to all the subject supervision data are matched and identified with the current supervision problem data in sequence, and the number of supervision problem data that meets the preset matching and identification requirements is used as the number of similar case supervision times data.

[0078] It should be noted that the supervisory subject is evaluated based on its historical supervisory data to assist in decision-making when the company handles the matter. The supervisory problem data corresponding to all subject supervisory data are matched and identified with the current supervisory problem data in turn. The number of supervisory problem data that meets the preset matching and identification requirements is used as similar case supervision frequency data. Similar case supervision frequency data indicates the number of times similar supervisory problems occur for the same supervisory subject.

[0079] According to an embodiment of the present invention, it also includes:

[0080] Extracting problem severity data and problem impact data based on the subject supervision data;

[0081] Calculate based on the problem severity data, problem impact data, supervision frequency data, and similar case supervision frequency data to obtain the supervision subject evaluation index;

[0082] The calculation formula of the supervisory subject evaluation index is:

[0083] ;

[0084] in, To monitor the subject evaluation index, For problem severity data, For the problem impact data, is the number of supervision data, is the number of similar case supervision data, , , , and is the preset characteristic coefficient (which can be obtained by querying the preset supervision problem rectification intelligent analysis platform);

[0085] The supervisory subject evaluation index is compared with a preset supervisory subject evaluation index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the supervisory subject is marked and a reminder message is sent.

[0086] It should be noted that in the embodiment of the present application, when the evaluation index of the supervisory subject exceeds a specific threshold, it indicates that there are serious problems with the supervisory subject. The supervisory subject needs to be marked and a reminder message needs to be sent to the client to assist the company in making decisions when handling the situation.

[0087] According to an embodiment of the present invention, it also includes:

[0088] Obtain the number of supervised problem data in the historical supervised data cluster, compare the number of supervised problem data with the number of supervised problem data threshold, and use the historical supervised data cluster corresponding to the number of supervised problem data that meets the preset threshold comparison requirement as the target supervised data cluster;

[0089] Using the supervised problem data corresponding to the target supervised data cluster as target supervised problem data;

[0090] Obtain data on the number of problem supervisors, problem frequency data, problem impact scope data, and problem controversy degree data of the target supervision problem data;

[0091] Calculate the company's operational impact index based on the problem supervision data, problem frequency data, problem impact scope data and problem controversy level data;

[0092] The calculation formula of the company's operational impact index is:

[0093] ;

[0094] in, The company's operational impact index. The number of supervisors for the problem is is the problem frequency data, For the problem impact range data, Data on the controversy level of the issue, , , and is the preset characteristic coefficient (which can be obtained by querying the preset supervision problem rectification intelligent analysis platform);

[0095] The company operation impact index is compared with a preset company operation impact index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the target supervision problem data is marked.

[0096] It should be noted that in the embodiment of the present application, the number of different historical supervision issues is compared with the problem number threshold, and the supervision issues with a large number of issues are taken as target supervision issues. The company operation impact index is obtained based on the problem supervision number data, problem frequency data, problem impact range data and problem dispute degree data of the target supervision issues. The company operation impact index is compared with the preset company operation impact index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirements, it means that the supervision issue has a large impact on the company's operations and needs to be marked for key rectification.

[0097] The present invention also discloses a supervision problem rectification intelligent analysis system based on a large model, comprising a memory and a processor, wherein the memory stores a supervision problem rectification intelligent analysis method program based on a large model, and when the supervision problem rectification intelligent analysis method program based on a large model is executed by the processor, the following steps are implemented:

[0098] Obtain historical supervision data and classify it, and insert class labels into the classified historical supervision data to obtain a class label data set;

[0099] Obtaining current supervision data and performing feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data;

[0100] A rectification strategy prediction model is obtained according to the target historical supervision data processing, the current feature data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client.

[0101] It should be noted that first, a rectification strategy prediction model is trained based on historical cases of similar supervision issues that have been resolved in the past. In the rectification phase of supervision issues, the rectification strategy prediction model is used to intelligently recommend supervision rectification methods, and the supervision subject is evaluated based on the historical supervision data of the supervision subject, so as to assist the company in making decisions when handling it. Finally, the target supervision issues with greater impact on the company's operations are screened out based on the evaluation of the impact of different supervision issues on the company's operations. This will achieve intelligent analysis of the rectification of supervision issues and improve supervision efficiency and company management level.

[0102] According to an embodiment of the present invention, the acquiring and classifying historical supervision data, and inserting class labels into the classified historical supervision data to obtain a class label data set includes:

[0103] Obtain historical supervision data and its corresponding company-level data, identify features of historical supervision data, and obtain feature data, including supervision subject data and supervision problem data;

[0104] The historical supervision data is classified according to the supervision problem data to obtain historical supervision data clusters of different categories, and class labels are inserted into the historical supervision data clusters to obtain a class label data set.

[0105] It should be noted that historical supervision data at different levels are classified and inserted into class labels according to the supervision problem to facilitate subsequent retrieval.

[0106] According to an embodiment of the present invention, the current supervision data is obtained and feature extraction is performed to obtain the current feature data, the current feature data is input into the class label data set for matching and identification, and the target historical supervision data is obtained by processing, including:

[0107] Obtain current supervision data and its corresponding current company-level data, perform feature extraction on the current supervision data, and obtain current feature data, including current supervision subject data and current supervision problem data;

[0108] The current supervision problem data is input into the class label data set for matching and identification, and the historical supervision data cluster corresponding to the class label that meets the preset matching and identification result is used as the target historical supervision data cluster;

[0109] The current company-level data is compared with the company-level data corresponding to all historical supervision data in the target historical supervision data cluster in sequence, and the historical supervision data corresponding to the company-level data that meets the preset comparison requirements is used as the target historical supervision data.

[0110] It should be noted that the current supervision problem data is extracted based on the current supervision data and input into the class label data set for matching and identification to obtain historical supervision data of the same type of supervision problem, and locate the target historical supervision data based on the company level where the current supervision data is located.

[0111] According to an embodiment of the present invention, the process of obtaining a rectification strategy prediction model based on the target historical supervision data, inputting the current feature data into the rectification strategy prediction model for processing, obtaining recommended rectification strategy data and pushing it to the client includes:

[0112] Extract problem rectification strategy data from the target historical supervision data, generate a supervision problem sample set from the supervision problem data set corresponding to all the target historical supervision data, and generate a problem rectification strategy sample set from the problem rectification strategy data set corresponding to all the target historical supervision data;

[0113] Using a preset machine learning algorithm to train the supervised problem sample set and the problem rectification strategy sample set to obtain a rectification strategy prediction model;

[0114] The current supervision problem data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client.

[0115] It should be noted that the supervision problem data set corresponding to all target historical supervision data is used to generate a supervision problem sample set, and the problem rectification strategy data set corresponding to all target historical supervision data is used to generate a problem rectification strategy sample set. Training is performed based on the supervision problem sample set and the problem rectification strategy sample set to obtain a rectification strategy prediction model, and the current supervision problem data is input into the rectification strategy prediction model for processing to obtain the recommended rectification strategy and push it to the client.

[0116] According to an embodiment of the present invention, it also includes:

[0117] Matching and identifying the current supervision subject data with all the supervision subject data in sequence, taking the historical supervision data corresponding to the supervision subject data that meets the preset matching and identification requirements as the subject supervision data, and taking the number of supervision subjects that meet the preset matching and identification requirements as the supervision frequency data;

[0118] The supervision problem data corresponding to all the subject supervision data are matched and identified with the current supervision problem data in sequence, and the number of supervision problem data that meets the preset matching and identification requirements is used as the number of similar case supervision times data.

[0119] It should be noted that the supervisory subject is evaluated based on its historical supervisory data to assist in decision-making when the company handles the matter. The supervisory problem data corresponding to all subject supervisory data are matched and identified with the current supervisory problem data in turn. The number of supervisory problem data that meets the preset matching and identification requirements is used as similar case supervision frequency data. Similar case supervision frequency data indicates the number of times similar supervisory problems occur for the same supervisory subject.

[0120] According to an embodiment of the present invention, it also includes:

[0121] Extracting problem severity data and problem impact data based on the subject supervision data;

[0122] Calculate based on the problem severity data, problem impact data, supervision frequency data, and similar case supervision frequency data to obtain the supervision subject evaluation index;

[0123] The calculation formula of the supervisory subject evaluation index is:

[0124] ;

[0125] in, To monitor the subject evaluation index, For problem severity data, For the problem impact data, is the number of supervision data, is the number of similar case supervision data, , , , and is the preset characteristic coefficient (which can be obtained by querying the preset supervision problem rectification intelligent analysis platform);

[0126] The supervisory subject evaluation index is compared with a preset supervisory subject evaluation index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the supervisory subject is marked and a reminder message is sent.

[0127] It should be noted that in the embodiment of the present application, when the evaluation index of the supervisory subject exceeds a specific threshold, it indicates that there are serious problems with the supervisory subject. The supervisory subject needs to be marked and a reminder message needs to be sent to the client to assist the company in making decisions when handling the situation.

[0128] According to an embodiment of the present invention, it also includes:

[0129] Obtain the number of supervised problem data in the historical supervised data cluster, compare the number of supervised problem data with the number of supervised problem data threshold, and use the historical supervised data cluster corresponding to the number of supervised problem data that meets the preset threshold comparison requirement as the target supervised data cluster;

[0130] Using the supervised problem data corresponding to the target supervised data cluster as target supervised problem data;

[0131] Obtain data on the number of problem supervisors, problem frequency data, problem impact scope data, and problem controversy degree data of the target supervision problem data;

[0132] Calculate the company's operational impact index based on the problem supervision data, problem frequency data, problem impact scope data and problem controversy level data;

[0133] The calculation formula of the company's operational impact index is:

[0134] ;

[0135] in, The company's operational impact index. The number of supervisors for the problem is is the problem frequency data, For the problem impact range data, Data on the controversy level of the issue, , , and is the preset characteristic coefficient (which can be obtained by querying the preset supervision problem rectification intelligent analysis platform);

[0136] The company operation impact index is compared with a preset company operation impact index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the target supervision problem data is marked.

[0137] It should be noted that in the embodiment of the present application, the number of different historical supervision issues is compared with the problem number threshold, and the supervision issues with a large number of issues are taken as target supervision issues. The company operation impact index is obtained based on the problem supervision number data, problem frequency data, problem impact range data and problem dispute degree data of the target supervision issues. The company operation impact index is compared with the preset company operation impact index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirements, it means that the supervision issue has a large impact on the company's operations and needs to be marked for key rectification.

[0138] The third aspect of the present invention provides a readable storage medium, which stores a program for an intelligent analysis method for rectification of supervised problems based on a large model. When the program for the intelligent analysis method for rectification of supervised problems based on a large model is executed by a processor, the steps of the intelligent analysis method for rectification of supervised problems based on a large model as described in any one of the above items are implemented.

[0139] The intelligent analysis method, system and medium for supervision problem rectification based on a large model disclosed in the present invention first train a rectification strategy prediction model based on historical cases of similar supervision problems that have been completed in the past. In the supervision problem rectification link, the rectification strategy prediction model is used to intelligently recommend supervision rectification methods, and the supervision subject is evaluated based on the historical supervision data of the supervision subject, so as to assist the company in making decisions when handling it. Finally, the target supervision problems with greater impact on the company's operations are screened out based on the evaluation of the impact of different supervision problems on the company's operations. In this way, intelligent analysis of supervision problem rectification is realized, and supervision efficiency and company management level are improved.

[0140] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0141] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0142] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0143] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, the aforementioned program can be stored in a readable storage medium, and when the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0144] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

Claims

1. An intelligent analysis method for supervision problem rectification based on a large model, characterized in that: The following steps are involved: Obtain historical supervision data and classify it, and insert class labels into the classified historical supervision data to obtain a class label data set; Obtaining current supervision data and performing feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data; Obtaining a rectification strategy prediction model based on the target historical supervision data, inputting the current feature data into the rectification strategy prediction model for processing, obtaining recommended rectification strategy data and pushing it to the client; The obtaining and classifying the historical supervision data, and inserting class labels into the classified historical supervision data to obtain a class label data set, includes: Obtain historical supervision data and its corresponding company-level data, identify features of historical supervision data, and obtain feature data, including supervision subject data and supervision problem data; Classifying the historical supervision data according to the supervision problem data to obtain historical supervision data clusters of different categories, and inserting class labels into the historical supervision data clusters to obtain class label data sets; The obtaining of current supervision data and feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data, includes: Obtain current supervision data and its corresponding current company-level data, perform feature extraction on the current supervision data, and obtain current feature data, including current supervision subject data and current supervision problem data; The current supervision problem data is input into the class label data set for matching and identification, and the historical supervision data cluster corresponding to the class label that meets the preset matching and identification result is used as the target historical supervision data cluster; Compare the current company-level data with the company-level data corresponding to all historical supervision data in the target historical supervision data cluster in sequence, and use the historical supervision data corresponding to the company-level data that meets the preset comparison requirements as the target historical supervision data; Matching and identifying the current supervision subject data with all the supervision subject data in sequence, taking the historical supervision data corresponding to the supervision subject data that meets the preset matching and identification requirements as the subject supervision data, and taking the number of supervision subjects that meet the preset matching and identification requirements as the supervision frequency data; Matching and identifying the supervision problem data corresponding to all the subject supervision data with the current supervision problem data in sequence, and taking the number of supervision problem data that meets the preset matching and identification requirements as the number of similar case supervision times data; Extracting problem severity data and problem impact data based on the subject supervision data; Calculate based on the problem severity data, problem impact data, supervision frequency data, and similar case supervision frequency data to obtain the supervision subject evaluation index; The calculation formula of the supervisory subject evaluation index is: ； in, To monitor the subject evaluation index, For problem severity data, For the problem impact data, is the number of supervision data, is the number of similar case supervision data, , , , and is the preset characteristic coefficient; The supervisory subject evaluation index is compared with a preset supervisory subject evaluation index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the supervisory subject is marked and a reminder message is sent.

2. The intelligent analysis method for supervision problem rectification based on a large model according to claim 1 is characterized in that: The step of obtaining a rectification strategy prediction model based on the target historical supervision data, inputting the current feature data into the rectification strategy prediction model for processing, obtaining recommended rectification strategy data and pushing it to the client includes: Extract problem rectification strategy data from the target historical supervision data, generate a supervision problem sample set from the supervision problem data set corresponding to all the target historical supervision data, and generate a problem rectification strategy sample set from the problem rectification strategy data set corresponding to all the target historical supervision data; Using a preset machine learning algorithm to train the supervised problem sample set and the problem rectification strategy sample set to obtain a rectification strategy prediction model; The current supervision problem data is input into the rectification strategy prediction model for processing, and the recommended rectification strategy data is obtained and pushed to the client.

3. The intelligent analysis method for supervision problem rectification based on a large model according to claim 2 is characterized in that: Also includes: Obtain the number of supervised problem data in the historical supervised data cluster, compare the number of supervised problem data with a threshold value of the number of supervised problem data, and use the historical supervised data cluster corresponding to the number of supervised problem data that meets the preset threshold comparison requirement as the target supervised data cluster; Using the supervised problem data corresponding to the target supervised data cluster as target supervised problem data; Obtain data on the number of problem supervisors, problem frequency data, problem impact scope data, and problem controversy degree data of the target supervision problem data; Calculate the company's operational impact index based on the problem supervision data, problem frequency data, problem impact scope data and problem controversy level data; The calculation formula of the company's operational impact index is: ； in, The company's operational impact index. The number of supervisors for the problem is is the problem frequency data, For the problem impact range data, Data on the controversy level of the issue, , , and is the preset characteristic coefficient; The company operation impact index is compared with a preset company operation impact index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the target supervision problem data is marked.

4. An intelligent analysis system for supervision problem rectification based on a large model, characterized by: The invention comprises a memory and a processor, wherein the memory stores a program of a method for intelligent analysis of supervision problem rectification based on a large model, and when the method program for intelligent analysis of supervision problem rectification based on a large model is executed by the processor, the following steps are implemented: Obtain historical supervision data and classify it, and insert class labels into the classified historical supervision data to obtain a class label data set; Obtaining current supervision data and performing feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data; Obtaining a rectification strategy prediction model based on the target historical supervision data, inputting the current feature data into the rectification strategy prediction model for processing, obtaining recommended rectification strategy data and pushing it to the client; The obtaining and classifying the historical supervision data, and inserting class labels into the classified historical supervision data to obtain a class label data set, includes: Obtain historical supervision data and its corresponding company-level data, identify features of historical supervision data, and obtain feature data, including supervision subject data and supervision problem data; Classifying the historical supervision data according to the supervision problem data to obtain historical supervision data clusters of different categories, and inserting class labels into the historical supervision data clusters to obtain class label data sets; The obtaining of current supervision data and feature extraction to obtain current feature data, inputting the current feature data into the class label data set for matching and identification, and processing to obtain target historical supervision data, includes: Obtain current supervision data and its corresponding current company-level data, perform feature extraction on the current supervision data, and obtain current feature data, including current supervision subject data and current supervision problem data; The current supervision problem data is input into the class label data set for matching and identification, and the historical supervision data cluster corresponding to the class label that meets the preset matching and identification result is used as the target historical supervision data cluster; Compare the current company-level data with the company-level data corresponding to all historical supervision data in the target historical supervision data cluster in sequence, and use the historical supervision data corresponding to the company-level data that meets the preset comparison requirements as the target historical supervision data; Matching and identifying the current supervision subject data with all the supervision subject data in sequence, taking the historical supervision data corresponding to the supervision subject data that meets the preset matching and identification requirements as the subject supervision data, and taking the number of supervision subjects that meet the preset matching and identification requirements as the supervision frequency data; Matching and identifying the supervision problem data corresponding to all the subject supervision data with the current supervision problem data in sequence, and taking the number of supervision problem data that meets the preset matching and identification requirements as the number of similar case supervision times data; Extracting problem severity data and problem impact data based on the subject supervision data; Calculate based on the problem severity data, problem impact data, supervision frequency data, and similar case supervision frequency data to obtain the supervision subject evaluation index; The calculation formula of the supervisory subject evaluation index is: ； in, To monitor the subject evaluation index, For problem severity data, For the problem impact data, is the number of supervision data, is the number of similar case supervision data, , , , and is the preset characteristic coefficient; The supervisory subject evaluation index is compared with a preset supervisory subject evaluation index threshold. If the threshold comparison result does not meet the preset threshold comparison result requirement, the supervisory subject is marked and a reminder message is sent.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a large-model-based intelligent analysis program for the rectification of supervised problems. When the large-model-based intelligent analysis program for the rectification of supervised problems is executed by a processor, the steps of the large-model-based intelligent analysis method for the rectification of supervised problems are implemented as described in any one of claims 1 to 3.

Images