Multi-dimensional salary debt early warning method and multi-dimensional salary debt early warning system

Abstract

The invention provides a multi-dimensional salary debt early warning method and system, and relates to the technical field of computer application, and the specific scheme comprises the steps: obtaining the multi-dimensional information of a target enterprise, and carrying out the preprocessing; extracting feature vectors in the multi-dimensional information; inputting the extracted feature vectors into a trained early warning model, and carrying out debt judgment based on a multi-dimensional index to obtain a risk level; based on the risk level, carrying out early warning on the risk that the target enterprise is lack of salaries; wherein the multi-dimensional information comprises an enterprise profit growth condition, an enterprise water and electricity consumption condition, an enterprise tax payment condition, an enterprise recruitment and dictionary condition, a debt operation condition, an enterprise labor and capital dispute condition and an enterprise credit evaluation condition; according to the method, the multi-dimensional information of the target enterprise is fused, and through intelligent multi-dimensional index prediction and salary debt judgment, the accuracy of salary debt early warning is improved.

Description

Technical Field

[0001] The present invention belongs to the field of computer application technology, and in particular relates to a multi-dimensional wage arrears warning method and system. Background Art

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

[0003] As the rights and interests of key groups of migrant workers are increasingly valued, special actions are being launched to provide services and protection for migrant workers, employment dynamics of migrant workers are being monitored more closely, and requirements are being strengthened to prevent wage arrears for migrant workers and provide early warning of risks, and to improve long-term mechanisms to eradicate wage arrears.

[0004] At present, the methods for monitoring and preventing wage arrears are relatively limited. Most rely on manual inspections by government departments or intervention by trade unions, which is usually not timely and effective. Although some intelligent early warning methods have emerged, they are easily affected by the quality of data selection, resulting in uneven early warning effects. Summary of the invention

[0005] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a multi-dimensional wage arrears warning method and system, which integrates the multi-dimensional information of the target enterprise and improves the accuracy of the wage arrears warning through intelligent multi-dimensional indicator prediction and wage arrears judgment.

[0006] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0007] A first aspect of the present invention provides a multi-dimensional wage arrears warning method.

[0008] A multi-dimensional wage arrears early warning method, including:

[0009] Obtain multi-dimensional information of the target enterprise and perform pre-processing;

[0010] Extract feature vectors from multi-dimensional information;

[0011] The extracted feature vector is input into the trained early warning model to make wage arrears judgment based on multi-dimensional indicators and obtain the risk level;

[0012] Based on the risk level, early warning is issued for the risk of wage arrears of the target enterprise;

[0013] Among them, the multi-dimensional information includes enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt operation, enterprise labor disputes, and enterprise credit assessment.

[0014] Furthermore, the preprocessing includes cleaning, denoising and formatting.

[0015] Furthermore, the feature vector is extracted by using the word2vec algorithm to convert the preprocessed multi-dimensional information into a vector matrix to achieve vectorized representation of the text data.

[0016] Furthermore, the early warning model includes prediction of multi-dimensional indicator values ​​and wage arrears judgment;

[0017] The prediction of the multi-dimensional index value is to use the extracted feature vector to predict the enterprise stability index, enterprise normal operation index, attendance abnormality index, unit wage arrears index, labor relationship index, and credit index;

[0018] The wage arrears judgment is to obtain the risk level by comparing the prediction result with the preset threshold.

[0019] Furthermore, the comparison between the prediction result and the preset threshold value is specifically:

[0020] By comparing the predicted value of the indicator with the first threshold, the score of each indicator is obtained;

[0021] The scores of each indicator are weighted averaged and compared with the second threshold to obtain the risk level.

[0022] Furthermore, the training of the early warning model includes:

[0023] Pre-train the early warning model through the training data set;

[0024] The fine-tuning method is used to adjust and optimize the pre-trained early warning model in real time.

[0025] Furthermore, the early warning of the risk of wage arrears of the target enterprise is to send early warning alarm information corresponding to the risk level to the regulatory department and the wage recipient himself.

[0026] A second aspect of the present invention provides a multi-dimensional wage arrears warning system.

[0027] A multi-dimensional wage arrears warning system, including an information acquisition module, a feature extraction module, a wage arrears judgment module and a risk warning module:

[0028] The information acquisition module is configured to: acquire multi-dimensional information of the target enterprise and perform pre-processing;

[0029] The feature extraction module is configured to: extract feature vectors from the multi-dimensional information;

[0030] The wage arrears judgment module is configured to: input the extracted feature vector into the trained early warning model, make wage arrears judgment based on multi-dimensional indicators, and obtain the risk level;

[0031] The risk warning module is configured to: warn the target enterprise of the risk of wage arrears based on the risk level;

[0032] Among them, the multi-dimensional information includes enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt operation, enterprise labor disputes, and enterprise credit assessment.

[0033] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of a multi-dimensional wage arrears warning method as described in the first aspect of the present invention.

[0034] The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a multi-dimensional wage arrears warning method as described in the first aspect of the present invention are implemented.

[0035] One or more of the above technical solutions have the following beneficial effects:

[0036] The present invention is based on multi-dimensional information such as enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt management, enterprise labor disputes, and enterprise credit evaluation. It comprehensively considers various aspects of the target enterprise and uses an intelligent method to predict the values ​​of multi-dimensional indicators. It uses multi-dimensional indicators to issue wage arrears warnings to improve the accuracy of wage arrears warnings.

[0037] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0039] Figure 1 This is a flow chart of the method of the first embodiment. DETAILED DESCRIPTION

[0040] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0041] 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 application. 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 "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0042] Embodiment 1

[0043] In one embodiment of the present disclosure, a multi-dimensional wage arrears warning method is provided, such as Figure 1 As shown, the following steps are included:

[0044] Step S1: Obtain multi-dimensional information of the target enterprise and perform preprocessing.

[0045] Among them, the multi-dimensional information includes enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt operation, enterprise labor disputes, and enterprise credit assessment.

[0046] Furthermore, the preprocessing includes cleaning, denoising and formatting.

[0047] Specifically, it integrates and manages real-name data, and at the same time brings together comprehensive information from multiple departments such as human resources and social security, public security, justice, finance, housing and construction, water conservancy, and power supply, including but not limited to corporate profit growth, water and electricity usage, tax payment, recruitment and resignation records, debt status, labor-management disputes, credit assessment, etc.

[0048] In order to improve the accuracy of the early warning, the arrears information, debt information, and dispute information in the above data are extracted separately and merged into multi-dimensional information to ensure the accuracy and completeness of the data.

[0049] Through data cleaning, deduplication, verification and other management measures, we improve data quality, build a dynamic data resource library, and establish a wage payment warning library based on this, so as to realize the visual presentation of warning scenarios and ensure the integrity, accuracy and availability of data.

[0050] Step S2: Extract feature vectors from multi-dimensional information.

[0051] Furthermore, the feature vector is extracted by using the word2vec algorithm to convert the preprocessed multi-dimensional information into a vector matrix to achieve vector representation of text data. Specifically, the social security arrears data (name, ID number, arrears amount, arrears month, etc.) is converted into a vector, and the multi-dimensional data is converted into a vector matrix together.

[0052] Step S3: Input the extracted feature vector into the trained early warning model to make wage arrears judgment based on multi-dimensional indicators to obtain the risk level.

[0053] Furthermore, the early warning model includes prediction of multi-dimensional indicator values ​​and wage arrears judgment;

[0054] The prediction of the multi-dimensional index value is to use the extracted feature vector to predict the enterprise stability index, enterprise normal operation index, attendance abnormality index, unit wage arrears index, labor relationship index, and credit index;

[0055] The wage arrears judgment is to obtain the risk level by comparing the prediction result with the preset threshold.

[0056] Furthermore, the comparison between the prediction result and the preset threshold value is specifically:

[0057] (1) Obtaining the score of each indicator by comparing the predicted value of the indicator with the first threshold;

[0058] The first threshold includes a plurality of threshold intervals, and the score is determined by judging in which interval the predicted value of the indicator is.

[0059] (2) Take a weighted average of the scores of each indicator and compare them with the second threshold to obtain the risk level.

[0060] The second threshold includes multiple threshold intervals, and the risk level is determined by judging in which interval the weighted average comprehensive score is.

[0061] The risk levels include 0-3 levels, which correspond to pre-set threshold intervals. For example, the weighted average comprehensive score between 0-25 is level 0, which means there is no risk of wage arrears; between 25-50 is level 1; between 50-75 is level 2; and between 75-100 is level 3. The higher the level, the greater the risk of wage arrears.

[0062] Specifically, machine learning or deep learning technologies, such as vector machines, decision trees, covariance, GBDT, SVN, and CNN, are used to analyze and model the extracted features related to wage arrears warning, and to build and train warning models to automatically identify and analyze warning indicators for wage payment.

[0063] This embodiment provides an early warning model based on a vector machine, and the construction process is as follows:

[0064] (1) Correlation analysis and processing of feature data

[0065] The linear relationship between each feature is evaluated using the covariance and other correlation coefficient methods, and the strength and direction of the linear relationship between each feature is evaluated to identify the influence of each indicator on the wage payment warning level results. The greater the influence, the stronger the correlation. After the evaluation is completed, the redundant features are eliminated and the initial parameters of the model are determined according to the degree of correlation.

[0066] (2) Based on the determined initial parameters, the early warning model is built using the vector machine algorithm. The feature data is divided into a training set and a test set. The training set data is used to train the vector machine model, and the test set data is used to evaluate the performance of the model.

[0067] (3) Training of early warning models: training the models and fine-tuning and evaluating the training results to achieve automatic identification and analysis of early warning indicators.

[0068] (4) Explanation and verification of the warning results: Based on the vector machine model, the best features are recursively selected and the data set is divided into a tree structure. At each node, the data is divided into different subsets according to the different feature values ​​until the stopping condition is met. The warning results are further explained and verified to provide risk levels and corresponding recommended treatment measures.

[0069] Furthermore, the training of the early warning model includes:

[0070] (1) Pre-train the early warning model using the training data set;

[0071] The pre-training uses the pre-processed multi-dimensional information as sample data to construct a training data set, and uses Python and Spark computing frameworks, combined with multiple optimization algorithms such as multi-party secure computing (MPC) and federated learning (FL), to achieve automation of data processing and modeling processes, including intelligent derivation of feature variables, intelligent adjustment of algorithm selection and parameter optimization, to ensure efficient and secure model training.

[0072] (2) Use fine-tuning methods to adjust and optimize the pre-trained warning model in real time.

[0073] Combined with efficient fine-tuning methods such as LoRA, PrefixTuning, P-Tuning, PromtTuning, AdaLoRA, etc., the pre-trained algorithm model is further optimized. Combined with real-time data feedback, the algorithm model is adjusted and optimized in real time to improve the accuracy and stability of the early warning model.

[0074] Specifically, the pre-trained model is fine-tuned and the weight parameters of the pre-trained model are frozen. While freezing the original model parameters, additional network layers are added to the model (such as fully connected layers, convolutional layers, etc. The fully connected layers are used to integrate the features obtained from the pre-trained model and output prediction results such as wage arrears warning information and risk level). Only these newly added network layer parameters are trained, and the model is optimized by fine-tuning to make it better adapt to the specific task of wage arrears warning.

[0075] After training is completed, multiple evaluation indicators such as ROC / AUC, KS, PSI, LIFT, etc. are used to regularly evaluate the model through indicators such as accuracy, recall rate, F1 value, etc., to conduct a comprehensive evaluation of the model performance to ensure that the model warning effect is optimal.

[0076] Step S4: Based on the risk level, an early warning is issued for the risk of wage arrears of the target enterprise.

[0077] Furthermore, the early warning of the risk of wage arrears of the target enterprise is to send early warning alarm information corresponding to the risk level to the regulatory department and the wage recipient himself.

[0078] Specifically, when the early warning model detects the risk of wage arrears, the system automatically triggers an early warning prompt. The early warning information can be sent to relevant government departments, trade union organizations and wage recipients through mobile phone applications, text messages, emails, etc. The early warning information may include risk levels, specific early warning indicators, recommended handling measures, etc., so that timely response measures can be taken to deal with the problem;

[0079] The early warning system can be linked with mobile law enforcement monitoring equipment to achieve real-time monitoring and inspection of employers involved in the risk of wage arrears. Law enforcement and monitoring personnel can receive early warning information through mobile terminals and conduct investigations and disposal work in a timely manner to protect the rights and interests of migrant workers;

[0080] In response to the risk of wage arrears, the system provides on-site disposal plans and operating guidelines to help law enforcement inspectors quickly deal with risk issues and cancel the corresponding alarm status after the problem is resolved.

[0081] Embodiment 2

[0082] In one embodiment of the present disclosure, a multi-dimensional wage arrears warning system is provided, including an information acquisition module, a feature extraction module, a wage arrears judgment module and a risk warning module:

[0083] The information acquisition module is configured to: acquire multi-dimensional information of the target enterprise and perform pre-processing;

[0084] The feature extraction module is configured to: extract feature vectors from the multi-dimensional information;

[0085] The wage arrears judgment module is configured to: input the extracted feature vector into the trained early warning model, make wage arrears judgment based on multi-dimensional indicators, and obtain the risk level;

[0086] The risk warning module is configured to: warn the target enterprise of the risk of wage arrears based on the risk level;

[0087] Among them, the multi-dimensional information includes enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt operation, enterprise labor disputes, and enterprise credit assessment.

[0088] Embodiment 3

[0089] The purpose of this embodiment is to provide a computer-readable storage medium.

[0090] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in a multi-dimensional wage arrears warning method as described in the first embodiment of the present disclosure.

[0091] Embodiment 4

[0092] The purpose of this embodiment is to provide an electronic device.

[0093] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in a multi-dimensional wage arrears warning method as described in the first embodiment of the present disclosure are implemented.

[0094] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-dimensional wage arrears warning method, characterized in that: include: Obtain multi-dimensional information of the target enterprise and perform pre-processing; Extract feature vectors from multi-dimensional information; The extracted feature vector is input into the trained early warning model to make wage arrears judgment based on multi-dimensional indicators and obtain the risk level; Based on the risk level, early warning is issued for the risk of wage arrears of the target enterprise; Among them, the multi-dimensional information includes enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt operation, enterprise labor disputes, and enterprise credit assessment.

2. A multi-dimensional wage arrears warning method as claimed in claim 1, characterized in that: The preprocessing includes cleaning, denoising and formatting.

3. A multi-dimensional wage arrears warning method as claimed in claim 1, characterized in that: The feature vector is extracted by using the word2vec algorithm to convert the pre-processed multi-dimensional information into a vector matrix to achieve vectorized representation of text data.

4. A multi-dimensional wage arrears warning method as claimed in claim 1, characterized in that: The early warning model includes prediction of multi-dimensional indicator values ​​and determination of wage arrears; The prediction of the multi-dimensional index value is to use the extracted feature vector to predict the enterprise stability index, enterprise normal operation index, attendance abnormality index, unit wage arrears index, labor relationship index, and credit index; The wage arrears judgment is to obtain the risk level by comparing the prediction result with the preset threshold.

5. A multi-dimensional wage arrears warning method as claimed in claim 4, characterized in that: The comparison between the prediction result and the preset threshold is specifically: By comparing the predicted value of the indicator with the first threshold, the score of each indicator is obtained; The scores of each indicator are weighted averaged and compared with the second threshold to obtain the risk level.

6. A multi-dimensional wage arrears warning method as claimed in claim 1, characterized in that: The training of the early warning model includes: Pre-train the early warning model through the training data set; The fine-tuning method is used to adjust and optimize the pre-trained early warning model in real time.

7. A multi-dimensional wage arrears warning method as claimed in claim 1, characterized in that: The aforementioned early warning of the risk of wage arrears of the target enterprise is to send early warning alarm information corresponding to the risk level to the regulatory department and the wage recipient himself.

8. A multi-dimensional wage arrears warning system, characterized in that: It includes information acquisition module, feature extraction module, wage arrears judgment module and risk warning module: The information acquisition module is configured to: acquire multi-dimensional information of the target enterprise and perform pre-processing; The feature extraction module is configured to: extract feature vectors from the multi-dimensional information; The wage arrears judgment module is configured to: input the extracted feature vector into the trained early warning model, make wage arrears judgment based on multi-dimensional indicators, and obtain the risk level; The risk warning module is configured to: warn the target enterprise of the risk of wage arrears based on the risk level; Among them, the multi-dimensional information includes enterprise profit growth, enterprise water and electricity consumption, enterprise tax payment, enterprise recruitment and resignation, debt operation, enterprise labor disputes, and enterprise credit assessment.

9. An electronic device, comprising: a memory for non-transitory storage of computer readable instructions; a processor for executing the computer-readable instructions; Wherein, when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 7 is executed.

10. A storage medium, characterized in that: The computer-readable instructions are non-transitorily stored, wherein when the computer-readable instructions are executed by a computer, the method of any one of claims 1 to 7 is performed.

Images