Unemployed insurance premium supervision method and system based on big data and similar population expansion

Abstract

The invention provides an unused insurance fund supervision method and system based on big data and similar crowd expansion, and relates to the technical field of unused insurance fund supervision, and the method comprises the steps: obtaining unused insurance fund related data, and constructing a data set; preprocessing the acquired unemployed insurance fund related data; on the basis of the preprocessed data, an unemployed insurance fund supervision model is constructed through a linear regression algorithm, and the unemployed insurance fund supervision model identifies an abnormal declaration target population on the basis of the preprocessed related data; based on the abnormal declaration target crowd output by the unemployed insurance fund supervision model, through a similar crowd expansion algorithm, constructing an unemployed insurance fund reverse supervision model, analyzing various features of the abnormal declaration target crowd, carrying out similarity calculation, calculating the distance between various feature vectors, and recognizing users similar to the abnormal declaration target crowd; and association mining of abnormal declarators-enterprises-batch abnormal declarators is realized, and subsequent supervision auditing is carried out.

Description

Technical Field

[0001] The present disclosure relates to the technical field of unemployment insurance fund supervision, and in particular to an unemployment insurance fund supervision method and system based on big data and similar population expansion. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] Unemployment insurance is an important part of the social security system, and the corresponding unemployment insurance fund plays a vital role. The unemployment insurance fund is a special fund established to deal with unemployment risks. It aims to provide financial assistance to eligible unemployed people and promote their re-employment through the accumulation and reasonable allocation of funds. The management and use of unemployment insurance funds follow strict regulations and procedures. The fund is uniformly managed by a special agency, and the funds are used for special purposes to ensure the safety and effective use of funds. When unemployed people meet the conditions for receiving unemployment insurance benefits, they can apply to relevant agencies. After review and confirmation, they can receive a certain amount of unemployment insurance benefits according to the time to ease economic pressure and ensure basic living. The effective management and strict supervision of unemployment insurance benefits are related to the vital interests of the majority of insured persons.

[0004] At present, there are problems in the management and use of unemployment insurance funds, such as single regulatory means and delayed risk warning, which creates opportunities for some illegal enterprises or individuals to defraud unemployment insurance benefits for people who do not meet the requirements for applying for unemployment insurance benefits by fabricating some reporting conditions and processes, resulting in the loss of unemployment insurance benefits. There are certain hidden dangers and loopholes in process supervision, and risk warnings cannot be issued in a timely manner. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes an unemployment insurance fund supervision method and system based on big data and similar population expansion. Based on massive data, an unemployment insurance fund supervision model is constructed through a linear regression algorithm, and an unemployment insurance fund reverse supervision model is constructed through a similar population expansion algorithm to accurately identify risk points and violations.

[0006] According to some embodiments, the present disclosure adopts the following technical solutions:

[0007] Unemployment insurance supervision system based on big data and similar population expansion, including:

[0008] The data collection module is configured to obtain unemployment insurance fund related data and construct a data set;

[0009] The data governance module is configured to pre-process the acquired unemployment insurance fund related data and use statistical methods to obtain data distribution and data correlation;

[0010] A supervision model building module is configured to build an unemployment insurance fund supervision model based on the preprocessed data through a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies a target population for abnormal reporting based on the preprocessed relevant data;

[0011] The reverse supervision model construction module is configured to build a reverse supervision model for the unemployment insurance fund based on the target population for abnormal reporting output by the unemployment insurance fund supervision model. It uses a similar population expansion algorithm to analyze various characteristics of the target population for abnormal reporting and perform similarity calculations. It calculates the distances between various characteristic vectors to identify users similar to the target population for abnormal reporting, and realizes the association mining between abnormal reporting persons, enterprises, and batch abnormal reporting persons, so as to conduct subsequent supervision and review.

[0012] According to some embodiments, the present disclosure adopts the following technical solutions:

[0013] The supervision method of the unemployment insurance supervision system based on big data and similar population expansion includes:

[0014] Obtain unemployment insurance fund related data and construct a data set;

[0015] Pre-process the obtained unemployment insurance fund related data and use statistical methods to obtain data distribution and data correlation;

[0016] Based on the pre-processed data, an unemployment insurance fund supervision model is constructed by a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies a target population for abnormal reporting based on the pre-processed relevant data;

[0017] Based on the target population of abnormal reporting output by the unemployment insurance fund supervision model, a reverse supervision model of the unemployment insurance fund is constructed through a similar population expansion algorithm. The various characteristics of the target population of abnormal reporting are analyzed and similarity is calculated. The distance between various feature vectors is calculated to identify users who are similar to the target population of abnormal reporting, and the association mining between abnormal reporting persons, enterprises and batch abnormal reporting persons is realized for subsequent supervision and review.

[0018] According to some embodiments, the present disclosure adopts the following technical solutions:

[0019] A non-transitory computer-readable storage medium, which is used to store computer instructions. When the computer instructions are executed by a processor, the supervision method of the unemployment insurance supervision system based on big data and similar population expansion is implemented.

[0020] According to some embodiments, the present disclosure adopts the following technical solutions:

[0021] An electronic device comprises: a processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes the supervision method of the unemployment insurance supervision system based on big data and similar population expansion.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention discloses an unemployment insurance fund supervision system based on big data and similar population expansion, which is supported by massive data and uses a linear regression algorithm as a core algorithm to build an unemployment insurance fund supervision model, realize intelligent early warning, quickly and accurately determine whether the applicant meets the conditions for receiving the benefits, and prevent the occurrence of fraud and false claims.

[0024] The present invention discloses an unemployment insurance fund supervision system based on big data and similar population expansion. Based on the similar population expansion algorithm lookalike, an unemployment insurance fund supervision model is constructed, various characteristics of abnormal applicants are analyzed, and people who are highly similar to given abnormal applicants are identified from similar enterprises to which they belong, and association mining between abnormal applicants, enterprises and batches of abnormal applicants is realized, so as to realize precise supervision, convert input data into vectors, calculate the distance between vectors to find users similar to seed users, evaluate the constructed model using a test data set, check the accuracy and stability of the model, enhance the pertinence of fund supervision, and more accurately identify risk points and violations based on the analysis of actual output abnormal results or data, so as to prevent and supervise the illegal fraudulent claims of groups involving enterprises.

[0025] The disclosed unemployment insurance fund supervision system is based on big data and similar population expansion, and the supervision decision model capability based on big data is deeply integrated with the business handling system. In the process of handling business, the risk situation of the applicant is warned in real time through the model capability open interface, which is a typical application of data and technology empowerment business, and realizes the monitoring of business in progress. It promotes the supervision of unemployment funds to develop in a more intelligent, precise and efficient direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure.

[0027] Figure 1 is a schematic diagram of a method according to an embodiment of the present disclosure;

[0028] Figure 2 The present invention is a flowchart of the implementation process of the embodiment of the present invention. DETAILED DESCRIPTION

[0029] The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

[0030] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0031] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0032] Example 1

[0033] In one embodiment of the present disclosure, an unemployment insurance fund supervision system based on big data and similar population expansion is provided, including:

[0034] The data collection module is configured to obtain unemployment insurance fund related data and construct a data set;

[0035] The data governance module is configured to pre-process the acquired unemployment insurance fund related data and use statistical methods to obtain data distribution and data correlation;

[0036] A supervision model building module is configured to build an unemployment insurance fund supervision model based on the preprocessed data through a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies a target population for abnormal reporting based on the preprocessed relevant data;

[0037] The reverse supervision model construction module is configured to build a reverse supervision model for the unemployment insurance fund based on the target population for abnormal reporting output by the unemployment insurance fund supervision model. It uses a similar population expansion algorithm to analyze various characteristics of the target population for abnormal reporting and perform similarity calculations. It calculates the distances between various characteristic vectors to identify users similar to the target population for abnormal reporting, and realizes the association mining between abnormal reporting persons, enterprises, and batch abnormal reporting persons, so as to conduct subsequent supervision and review.

[0038] As an embodiment, in the data collection module, data related to the unemployment insurance fund is obtained and a data set is constructed, including: collecting data related to the unemployment insurance fund through multiple channels, including payment records, application information, approval results, labor contracts, contract signing records, etc., as well as data from external departments and institutions, and integrating these data together to form a complete data set.

[0039] In the data governance module, the obtained unemployment insurance fund-related data is preprocessed, including: data cleaning of the obtained data, identification of obviously erroneous data in the data, and screening out missing values, outliers, and inconsistent formats. Find and delete duplicate records, and only retain unique entries to avoid bias in the analysis process. Conduct in-depth exploratory analysis of the data, including data lineage tracking to understand the origin and flow of data, and use statistical methods to deeply analyze the characteristics of the data, take the mean, median and mode of the unemployment benefit collection data, that is, the average unemployment benefit collection value, median collection value, etc., and use standard deviation and variance to calculate the degree of deviation of the data. Including its distribution form and the correlation between variables, analyze the data distribution and data correlation. Data correlation refers to the existence of a certain statistical relationship between data. The change of one data can predict or explain the change of another data.

[0040] In the insurance fund supervision model construction module, first of all, the business data of abnormal cases of unemployment insurance benefits were analyzed, and a total of six types of abnormal unemployment insurance fund claims were identified, namely: 01-Employed 02-Enjoyed retirement benefits 03-False affiliation 04-False labor contract 05-Deceased 06-Others.

[0041] Then, an unemployment insurance fund supervision model is constructed with the linear regression algorithm as the core algorithm. The data required for model training include two categories. One category is internal business data, including insurance data, which is used to identify whether the insurance has been participated in; labor contract data is used to identify the authenticity of contract data; benefit data is used to identify whether the pension benefits have been enjoyed; employment data is used to identify whether the re-employment has been made; the other category is data shared by external departments, including death data, which is used to identify whether the person has died; and prison data, which is used to identify whether the person is serving a sentence during the period of collection.

[0042] The model is trained based on the accumulated data, and the best fitting line is found by minimizing the sum of squared errors. The model parameters obtained through regression analysis are used to establish a prediction model and predict future data points. Cross-validation technology is used to optimize model parameters to avoid overfitting.

[0043] Furthermore, the construction of the unemployment insurance fund supervision model includes:

[0044] Based on the aggregated and pre-processed high-quality business sedimentation data, with the linear regression algorithm as the core algorithm, the unemployment insurance fund supervision model is constructed through the construction methods such as problem definition, data collection, model selection, feature engineering, model training and evaluation. The essence of this model is to build a linear relationship between the abnormal type of unemployment insurance and various business data. The aggregated data includes internal data of human resources and social security, insurance data, labor contract data, benefit data, employment data, etc., and external department data includes prison data, death data, etc. The target population of abnormal declaration is taken as the target variable, and various characteristic data around the declarant are taken as characteristic variables. The unemployment insurance fund supervision model is based on the assumption of a linear relationship between the target variable (abnormal declarant) and the characteristic variable (various characteristic data around the declarant), and the characteristic variables are linearly combined to predict the value of the target variable.

[0045] The trained model was evaluated using the test data set to check the prediction accuracy and generalization ability of the model. Mean square error (MSE), root mean square error (RMSE), R^2 score and other methods were used to evaluate the model. Cross-validation and grid search were used to find the optimal combination of model parameters.

[0046] Based on the data calculated by the unemployment insurance fund supervision model, the personnel information and abnormal types that may exist are given, and the corresponding supporting explanations are given and pushed to the business personnel for confirmation and review.

[0047] In the reverse supervision model construction module, based on the target population of abnormal reporting output by the unemployment insurance fund supervision model, a reverse supervision model of the unemployment insurance fund is constructed through a similar population expansion algorithm. The various characteristics of the target population of abnormal reporting are analyzed and similarity is calculated. The distance between various feature vectors is calculated to identify users similar to the target population of abnormal reporting, realize the association mining of abnormal reporting person-enterprise-batch abnormal reporting person, and conduct subsequent supervision and review.

[0048] Based on the data calculated by the reverse supervision model of the unemployment insurance fund, the personnel information and abnormal types that may exist are given, and the corresponding supporting explanations are given and pushed to the business personnel for confirmation and review.

[0049] Furthermore, the reverse supervision model of unemployment insurance funds includes:

[0050] Using the two types of abnormal applicants, 03-false affiliation and 04-false labor contract, output by the unemployment insurance fund supervision model, the unemployment insurance fund supervision model is constructed based on the similar population expansion algorithm lookalike. The construction process of the model includes data collection and preprocessing, seed user definition, user portrait construction, selection of existing similarity calculation methods, model training and expansion population generation. The core of this model is to find similar people from massive data based on the characteristics of seed users (known abnormal applicants). Analyze the various characteristics of abnormal applicants, identify people who are highly similar to the given abnormal applicants from similar companies, and realize the association mining of abnormal applicants-enterprises-batch abnormal applicants, so as to achieve precise supervision.

[0051] Convert the input data into vectors, calculate the distance between vectors to find users similar to the seed users, and compare the vectors based on the seed users' characteristics such as work unit, age, industry, etc. to find similar users. Use the test data set to evaluate the constructed model and check the accuracy and stability of the model. Tune the model based on the evaluation results, including adjusting algorithm parameters and optimizing feature selection.

[0052] Example 2

[0053] In one embodiment of the present disclosure, a supervision method of an unemployment insurance fund supervision system based on big data and similar population expansion is provided, comprising:

[0054] Step 1: Obtain relevant data on unemployment insurance funds and construct a data set;

[0055] Step 2: Preprocess the obtained unemployment insurance fund related data and use statistical methods to obtain data distribution and data correlation;

[0056] Step 3: Based on the pre-processed data, an unemployment insurance fund supervision model is constructed by a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies the target population of abnormal declarations based on the pre-processed relevant data;

[0057] Step 4: Based on the target population of abnormal reporting output by the unemployment insurance fund supervision model, a reverse supervision model of the unemployment insurance fund is constructed through a similar population expansion algorithm. The various characteristics of the target population of abnormal reporting are analyzed and similarity is calculated. The distance between various feature vectors is calculated to identify users similar to the target population of abnormal reporting, and the association mining between abnormal reporting person, enterprise and batch abnormal reporting person is realized to conduct subsequent supervision and review.

[0058] Specifically, in step 1, the unemployment insurance fund related data are obtained and a data set is constructed;

[0059] Based on the collection module S101, the ETL tool is used to extract the data of the human resources and social security business system (including social security system, employment system, and labor relations system) into the designated data table on T+1, and the shared data of external departments is regularly saved into the designated data table through the interface.

[0060] In step 2, the obtained unemployment insurance fund related data are preprocessed, and statistical methods are used to obtain data distribution and data correlation;

[0061] Specifically, based on the data preprocessing module S102, it preprocesses the collected data, including cleaning the data, identifying abnormal data and duplicate data, standardizing the data, ensuring the high quality and availability of the data, and vectorizing unstructured data (such as labor contracts) to provide data support for model construction.

[0062] In step 3, based on the pre-processed data, an unemployment insurance fund supervision model is constructed by a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies the target population of abnormal declarations based on the pre-processed relevant data;

[0063] Specifically, an unemployment insurance fund supervision model is constructed with linear regression algorithm as the core algorithm.

[0064] The construction of the unemployment insurance fund supervision model includes:

[0065] In the unemployment insurance fund supervision module S103, based on the aggregated and pre-processed high-quality business sedimentation data, the unemployment insurance fund supervision model is constructed with the linear regression algorithm as the core algorithm. The aggregated data includes internal data of human resources and social security, insurance data, labor contract data, benefit data, employment data, etc., and external department data includes prison data, death data, etc. The target population of abnormal declaration is taken as the target variable, and various characteristic data around the declarant are taken as characteristic variables. The unemployment insurance fund supervision model is based on the assumption of a linear relationship between the target variable (abnormal declarant) and the characteristic variable (various characteristic data around the declarant), and the characteristic variables are linearly combined to predict the value of the target variable.

[0066] The trained model was evaluated using the test data set to check the prediction accuracy and generalization ability of the model. Mean square error (MSE), root mean square error (RMSE), R^2 score and other methods were used to evaluate the model. Cross-validation and grid search were used to find the optimal combination of model parameters.

[0067] Based on the data calculated by the unemployment insurance fund supervision model, the personnel information and abnormal types that may exist are given, and the corresponding supporting explanations are given and pushed to the business personnel for confirmation and review.

[0068] Step 4: Based on the target population of abnormal reporting output by the unemployment insurance fund supervision model, a reverse supervision model of the unemployment insurance fund is constructed through a similar population expansion algorithm. The various characteristics of the target population of abnormal reporting are analyzed and similarity is calculated. The distance between various feature vectors is calculated to identify users similar to the target population of abnormal reporting, and the association mining between abnormal reporting person, enterprise and batch abnormal reporting person is realized to conduct subsequent supervision and review.

[0069] The reverse supervision module S104 of unemployment insurance funds takes the similar population expansion algorithm Lookalike as the core, and uses the characteristic value to portray the population portrait and find similar populations. The two types of abnormal declarations, 03-false affiliation and 04-false labor contracts, are mostly participated by enterprises. Through the abnormal declaration person-enterprise-same type of people, a reverse closed loop of supervision is formed to improve the pertinence of supervision.

[0070] Model tuning and training module S105: The unemployment insurance fund supervision model uses the test data set to evaluate the trained model, check the prediction accuracy and generalization ability of the model, use the mean square error (MSE), root mean square error (RMSE), R^2 score and other methods to evaluate the model, and use cross-validation, grid search and other methods to find the optimal model parameter combination. The unemployment insurance fund reverse supervision model uses the test data set to evaluate the constructed model and check the accuracy and stability of the model. Tune the model according to the evaluation results, including adjusting algorithm parameters and optimizing feature selection.

[0071] The unemployment insurance fund supervision result output module S106 provides two ways to use the model supervision capability. One is to perform batch supervision on completed businesses based on business sedimentation data after the fact. The other is to output the model capability to the business system and call the model capability API in real time during the business handling process to achieve process supervision. The two methods can be used in parallel.

[0072] The unemployment insurance fund reverse supervision module S107 is called for specific situations. For abnormal results such as 03-false affiliation and 04-false labor contract when calling the unemployment insurance fund supervision model, the unemployment insurance fund reverse supervision model is called again to further deepen the supervision, use the abnormal declarant as the seed, conduct reverse supervision, and find similar groups.

[0073] The unemployment insurance fund reverse supervision result output module S108 calls the unemployment insurance fund reverse supervision model to output the abnormal reporting results and the basis description.

[0074] The module S109 that connects with the core business system outputs the output results of the unemployment insurance fund supervision model and the unemployment insurance fund reverse supervision model to the business system supervision end, and the supervisor will review the results and conduct subsequent business processing.

[0075] Example 3

[0076] In one embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the supervision method of the unemployment insurance supervision system based on big data and similar population expansion is implemented.

[0077] Example 4

[0078] In one embodiment of the present disclosure, an electronic device is provided, including: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the supervision method of the unemployment insurance supervision system based on big data and similar population expansion.

[0079] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes 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.

[0080] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0081] Although the above describes the specific implementation methods of the present disclosure in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present disclosure. Technical personnel in the relevant field should understand that on the basis of the technical solution of the present disclosure, various modifications or variations that can be made by those skilled in the art without creative work are still within the scope of protection of the present disclosure.

Claims

1. Unemployment insurance supervision system based on big data and similar population expansion, characterized by: include: The data collection module is configured to obtain unemployment insurance fund related data and construct a data set; The data governance module is configured to pre-process the acquired unemployment insurance fund related data and use statistical methods to obtain data distribution and data correlation; A supervision model building module is configured to build an unemployment insurance fund supervision model based on the preprocessed data through a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies a target population for abnormal reporting based on the preprocessed relevant data; The reverse supervision model construction module is configured to build a reverse supervision model for the unemployment insurance fund based on the target population for abnormal reporting output by the unemployment insurance fund supervision model. It uses a similar population expansion algorithm to analyze various characteristics of the target population for abnormal reporting and perform similarity calculations. It calculates the distances between various characteristic vectors to identify users similar to the target population for abnormal reporting, and realizes the association mining between abnormal reporting persons, enterprises, and batch abnormal reporting persons, so as to conduct subsequent supervision and review.

2. The unemployment insurance fund supervision system based on big data and similar population expansion as claimed in claim 1, characterized in that: Obtain relevant data on unemployment insurance funds, build a data set, including payment records, application information, approval results, labor contract workers, contract records, and data from external departments and institutions, and integrate them into a complete data set.

3. The unemployment insurance fund supervision system based on big data and similar population expansion as claimed in claim 1, characterized in that: The obtained unemployment insurance fund-related data is preprocessed, including: data cleaning, identifying obviously erroneous data, screening out missing values, outliers and inconsistent formats, finding and deleting duplicate records, retaining only unique entries, and using statistical methods to deeply analyze the characteristics of the data, including its distribution pattern and the correlation between variables, and analyzing the data distribution and data correlation.

4. The unemployment insurance fund supervision system based on big data and similar population expansion as claimed in claim 1, characterized in that: Based on the preprocessed data, an unemployment insurance fund supervision model is constructed using a linear regression algorithm, including: analyzing the business data of abnormal cases of unemployment insurance benefits, and obtaining six types of unemployment insurance fund application situations, namely employed, enjoying retirement benefits, false affiliation, false labor contract, deceased, and others. An unemployment insurance fund supervision model is constructed using a linear regression algorithm, and the model is trained based on the preprocessed data. The best fitting line is found by minimizing the sum of squared errors. The model parameters are obtained through regression analysis, and the unemployment insurance fund supervision model is established to predict future data points.

5. The unemployment insurance fund supervision system based on big data and similar population expansion as claimed in claim 4, characterized in that: The unemployment insurance fund supervision model outputs possible abnormal personnel information and abnormal types, and the target population for abnormal reporting is false affiliation and false labor contracts.

6. The unemployment insurance fund supervision system based on big data and similar population expansion as claimed in claim 1, characterized in that: Based on the similar population expansion algorithm lookalike, an unemployment insurance fund supervision model is constructed to analyze the various characteristics of the target population for abnormal reporting. The input data is converted into vectors, and the distance between vectors is calculated to find users similar to seed users. Seed users are the target population for abnormal reporting.

7. The unemployment insurance fund supervision system based on big data and similar population expansion as claimed in claim 1, characterized in that: The target population of abnormal declarations is taken as the target variable, and various characteristic data around the declarants are taken as characteristic variables. The unemployment insurance fund supervision model is based on the assumption of a linear relationship between the target variable and the characteristic variable. The feature variables are linearly combined to predict the value of the target variable.

8. A supervision method for an unemployment insurance fund supervision system based on big data and similar population expansion according to any one of claims 1 to 7, characterized in that: include: Obtain unemployment insurance fund related data and construct a data set; Pre-process the obtained unemployment insurance fund related data and use statistical methods to obtain data distribution and data correlation; Based on the pre-processed data, an unemployment insurance fund supervision model is constructed by a linear regression algorithm, wherein the unemployment insurance fund supervision model identifies a target population for abnormal reporting based on the pre-processed relevant data; Based on the target population of abnormal reporting output by the unemployment insurance fund supervision model, a reverse supervision model of the unemployment insurance fund is constructed through a similar population expansion algorithm. The various characteristics of the target population of abnormal reporting are analyzed and similarity is calculated. The distance between various feature vectors is calculated to identify users who are similar to the target population of abnormal reporting, and the association mining between abnormal reporting persons, enterprises and batch abnormal reporting persons is realized for subsequent supervision and review.

9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by the processor, the supervision method of the unemployment insurance supervision system based on big data and similar population expansion as described in any one of claim 8 is implemented.

10. An electronic device, characterized in that: include: A processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes the supervision method of the unemployment insurance supervision system based on big data and similar population expansion as described in any one of claim 8.

Images