CN110414806B

CN110414806B - Employee risk early warning method and related device

Info

Publication number: CN110414806B
Application number: CN201910618947.1A
Authority: CN
Inventors: 刘宇晗; 朱旻; 王景昱; 吴英豪; 薛志强
Original assignee: Ping An Technology Shenzhen Co Ltd
Current assignee: Ping An Technology Shenzhen Co Ltd
Priority date: 2019-07-10
Filing date: 2019-07-10
Publication date: 2024-05-14
Anticipated expiration: 2039-07-10
Also published as: CN110414806A

Abstract

The embodiment of the invention discloses an employee risk early warning method and a related device. The method comprises the following steps: acquiring behavior data for recording any worker behavior from an employee database; determining whether the behavior data is risk behavior data based on the risk comparison information set; when determining that the behavior data is the risk behavior data based on any risk comparison information in the risk comparison information set, determining a type label corresponding to any risk comparison information, determining an employee type corresponding to the type label as an employee type of an employee corresponding to the behavior data, and determining the employee corresponding to the behavior data as a risk employee; determining the risk level of the risk staff; generating early warning information and sending the early warning information to a risk early warning platform to perform risk early warning on risk staff. By adopting the embodiment of the invention, the risk staff and the risk behavior data of the risk staff can be accurately determined, the risk early warning can be carried out on the determined risk staff, and the applicability is high.

Description

Employee risk early warning method and related device

Technical Field

The invention relates to the technical field of computers, in particular to an employee risk early warning method and a related device.

Background

Along with the continuous improvement of an enterprise employee management system, various enterprises can perform employee evaluation work through an enterprise internal employee management system so as to screen out risk employees with risk behaviors and make corresponding early warning precautions. In the traditional mode, a method of combining manpower and machinery is generally adopted, and risk staff is determined from the staff management system by carrying out preliminary screening on staff data in the staff management system and then based on a manual sampling and manual distinguishing mode or based on personal experience of early warning staff.

However, in the above conventional manner, the determination of the risk employee has strong subjectivity and instability, and the risk determination for all employees in the employee management system cannot be completed, which makes it difficult to accurately and comprehensively determine the risk employee from the employee management system.

Disclosure of Invention

The embodiment of the invention provides an employee risk early warning method and a related device, which can accurately determine risk employees and risk behavior data of the risk employees, can timely perform risk early warning on the determined risk employees, and are high in applicability.

In a first aspect, an embodiment of the present invention provides an employee risk early warning method, where the method includes:

Acquiring behavior data for recording any worker behavior from an employee database;

Determining whether the behavior data is risk behavior data based on the risk comparison information set;

when determining that the behavior data is risk behavior data based on any risk comparison information in the risk comparison information set, determining a type label corresponding to the any risk comparison information, determining an employee type corresponding to the type label as an employee type of an employee corresponding to the behavior data, and determining the employee corresponding to the behavior data as a risk employee;

Acquiring risk information of the behavior data, and determining the risk level of the risk staff based on the risk information;

And generating early warning information based on the employee type, the behavior data and the risk grade, and sending the early warning information to a risk early warning platform to perform risk early warning on the risk employee.

With reference to the first aspect, in one possible implementation manner, before the acquiring, from the employee database, behavior data that records any one of the employee behaviors, the method further includes:

Acquiring risk behavior sample data of at least two employee type risk employees from an employee database, wherein the risk employee of one employee type corresponds to the at least two risk behavior sample data, and the risk behavior sample data is used for recording one risk behavior;

and constructing a risk comparison information set based on risk behavior sample data of the risk employees of the at least two employee types.

With reference to the first aspect, in one possible implementation manner, the acquiring risk behavior sample data of risk employees of at least two employee types from the employee database includes:

Determining risk employees of at least two employee types from an employee database;

inputting risk behavior data of risk employees of various employee types into a preset nonlinear model, and obtaining confidence degrees of various risk behavior data based on the preset nonlinear model;

And determining risk behavior data with the confidence coefficient within a preset confidence coefficient interval as risk behavior sample data to obtain risk behavior sample data of risk employees of the at least two employee types.

With reference to the first aspect, in a possible implementation manner, the constructing the risk comparison information set based on risk behavior sample data of risk employees of the at least two employee types includes:

Determining a type label corresponding to each employee type based on risk behavior sample data corresponding to each employee type;

Processing the risk behavior sample data of the risk employees of the at least two employee types in a data format to obtain risk behavior sample data with a preset data format;

Constructing a piece of risk comparison information based on the type label corresponding to each employee type and risk behavior sample data with a preset data format to obtain a risk comparison information set at least comprising two pieces of risk comparison information;

wherein, one piece of risk comparison information in the risk comparison information set is used for determining whether one piece of behavior data of one employee type is risk behavior data.

With reference to the first aspect, in a possible implementation manner, the risk information includes a behavior feature value and/or a risk behavior feature word; the acquiring the risk information of the behavior data, and determining the risk level of the risk employee based on the risk information includes:

determining a behavior characteristic value of the behavior data, and determining the risk level of the risk employee based on the behavior characteristic value and a preset risk level table; and/or the number of the groups of groups,

Carrying out data analysis on the risk behavior data to convert the behavior data into risk behavior text;

carrying out semantic recognition on the risk behavior text to determine risk behavior feature words of the risk staff from the risk behavior text;

and determining the risk grade of the risk employee based on the risk behavior feature words and a preset risk keyword library.

With reference to the first aspect, in one possible implementation manner, the generating early warning information based on the employee type, the behavior data, and the risk level, and sending the early warning information to a risk early warning platform to perform risk early warning on the risk employee includes:

obtaining employee information of the risk employee from the employee database;

determining whether the risk level is greater than a preset risk level;

If the risk level is greater than a preset risk level, determining an early warning measure corresponding to the risk staff based on staff behaviors recorded by the behavior data, and if the risk level is not greater than the preset risk level, determining the preset early warning measure as an early warning measure corresponding to the risk staff;

Generating early warning information based on the employee information, the employee type and the early warning measures, and sending the early warning information to a risk early warning platform so that the risk early warning platform displays the employee information and the employee type of the risk employee, and carrying out risk early warning on the risk employee based on the early warning measures in the early warning information.

In a second aspect, an embodiment of the present invention provides an employee risk early-warning device, including:

the acquisition unit is used for acquiring behavior data for recording any worker behavior from the staff database;

The judging unit is used for determining whether the behavior data are risk behavior data or not based on the risk comparison information set;

The determining unit is used for determining a type label corresponding to any risk comparison information when determining that the behavior data is risk behavior data based on any risk comparison information in the risk comparison information set, determining an employee type corresponding to the type label as an employee type of an employee corresponding to the behavior data, and determining the employee corresponding to the behavior data as a risk employee;

The determining unit is used for acquiring risk information of the behavior data and determining the risk level of the risk staff based on the risk information;

And the early warning unit is used for generating early warning information based on the employee type, the behavior data and the risk grade, and sending the early warning information to a risk early warning platform so as to perform risk early warning on the risk employee.

With reference to the second aspect, in a possible implementation manner, the acquiring unit is further configured to:

the employee risk early warning device further comprises:

The construction unit is further used for constructing a risk comparison information set based on risk behavior sample data of risk employees of the at least two employee types.

With reference to the second aspect, in a possible implementation manner, the determining unit is configured to:

The employee risk early warning device comprises: the processing unit is used for carrying out data format processing on the risk behavior sample data of the risk employees of the at least two employee types so as to obtain risk behavior sample data with preset data formats;

The construction unit is configured to construct a piece of risk comparison information based on the type tag corresponding to each employee type and risk behavior sample data with a preset data format, so as to obtain a risk comparison information set including at least two pieces of risk comparison information;

With reference to the second aspect, in a possible implementation manner, the risk information includes a behavior feature value and/or a risk behavior feature word; the above-mentioned determination unit is used for:

determining a behavior characteristic value of the behavior data, and determining the risk level of the risk employee based on the behavior characteristic value and a preset risk level table;

The processing unit is used for carrying out data analysis on the risk behavior data so as to convert the behavior data into risk behavior text;

the processing unit is used for carrying out semantic recognition on the risk behavior text so as to determine risk behavior feature words of the risk staff from the risk behavior text;

The determining unit is configured to determine a risk level of the risk employee based on the risk behavior feature word and a preset risk keyword library.

With reference to the second aspect, in one possible implementation manner, the acquiring unit is configured to:

obtaining employee information of the risk employee from the employee database;

the judging unit is used for determining whether the risk level is greater than a preset risk level;

the determining unit is configured to determine, based on employee behaviors recorded by the behavior data, an early warning measure corresponding to the risk employee when the risk level is greater than a preset risk level, and determine, when the risk level is not greater than the preset risk level, the preset early warning measure as an early warning measure corresponding to the risk employee;

The early warning unit is used for generating early warning information based on the employee information, the employee type and the early warning measures and sending the early warning information to the risk early warning platform so that the risk early warning platform displays the employee information and the employee type of the risk employee, and performing risk early warning on the risk employee based on the early warning measures in the early warning information.

In a third aspect, an embodiment of the present invention provides a terminal, where the terminal includes a processor and a memory, and the processor and the memory are connected to each other. The memory is configured to store a computer program supporting the terminal to perform the method provided by the first aspect and/or any of the possible implementation manners of the first aspect, the computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method provided by the first aspect and/or any of the possible implementation manners of the first aspect.

In a fourth aspect, embodiments of the present invention provide a computer readable storage medium storing a computer program for execution by a processor to implement the method provided by the first aspect and/or any one of the possible implementation manners of the first aspect.

In the embodiment of the invention, whether the behavior data of any employee in the employee database is the risk behavior data can be determined based on the risk comparison information set, and then the risk employee and the employee type, employee information, risk level and other information of the risk employee are determined under the condition that the behavior data of any employee is determined to be the risk behavior data. Based on the implementation mode, real-time detection of all behavior data in the employee database can be realized, and risk behavior data and risk employees can be determined in time. Meanwhile, corresponding early warning measures are determined based on various information such as employee types, employee information and risk grades of risk employees, and the risk early warning platform is indicated to perform risk early warning on the risk employees, so that the situation of inaccurate early warning caused by manual operation or subjective factors is avoided, and the risk early warning system is good in flexibility and high in applicability.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for employee risk early warning provided by an embodiment of the present invention;

FIG. 2 is a flowchart of a method for constructing a risk comparison information set according to an embodiment of the present invention;

Fig. 3 is a schematic structural diagram of an employee risk early warning device according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a terminal according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The employee risk early warning method provided by the embodiment of the invention (for convenience of description, the method provided by the embodiment of the invention can be simply referred to as a method provided by the embodiment of the invention) can be applied to various enterprise employee management systems in various fields and various industries, such as a bank employee management system, an employee behavior analysis system and the like, and is not limited herein. The method provided by the embodiment of the invention can monitor the employee behavior data in the enterprise employee management system in real time, can timely determine the risk behavior data and the risk employees corresponding to the risk behavior data, and further can perform risk early warning on the risk employees, and has high flexibility and wide application range. For convenience of description, the method and related apparatus provided by the embodiments of the present invention will be described below with reference to fig. 1 to 4, by taking any one of the worker management systems of a certain target enterprise in a certain field and/or a certain industry as an example.

Referring to fig. 1, fig. 1 is a flowchart of an employee risk early warning method according to an embodiment of the present invention. The method provided by the embodiment of the invention can comprise the following steps S11-S16:

s11, acquiring behavior data for recording any worker behavior from the staff database.

In some possible implementations, all behavioral data of all employees in the employee management system may be monitored. For each piece of behavior data, behavior data for recording any one of the worker behaviors may be acquired from an employee database in the employee management system. The employee database may be an employee data set defined by the employee management system, or may be a storage space (mobile hard disk, cloud storage, etc.) connected to the employee management system for storing behavioral data generated by the enterprise employee, which is not limited herein. The data types, data sizes, data formats, and the like of the behavior data of each employee in the employee database are not limited herein. It should be noted that, the employee database stores not only the behavioral data of the employee, but also the non-behavioral data (such as identity data and attendance data) of the employee. Therefore, when the behavior data for recording any one of the staff activities is acquired from the staff database, the behavior data corresponding to any one of the staff needs to be screened and acquired from the staff database by a big data analysis and processing method or the like, and the specific data screening method is not limited herein.

And S12, determining whether the behavior data is risk behavior data or not based on the risk comparison information set.

After the behavior data for recording any one of the worker behaviors is acquired, it may be determined whether the behavior data is risk behavior data based on the risk comparison information set. Wherein, the risk comparison information set can be constructed before the behavior data is acquired. As shown in fig. 2, fig. 2 is a flowchart of a method for constructing a risk comparison information set according to an embodiment of the present invention, where the method for constructing a risk comparison information set according to an embodiment of the present invention includes steps S21 to S24 as follows:

s21, acquiring risk behavior sample data of risk employees of at least two employee types from an employee database.

In some possible embodiments, risk employees of at least two employee types may be determined from an employee database, and then risk behavioral profile data may be obtained from multiple risk data for the risk employees of at least two employee types. One employee type may correspond to a plurality of risk employees, one risk employee corresponds to at least two risk sample data, and one risk behavior data is used for recording one risk behavior. Since risk behavior data with weaker risk behaviors may be recorded in the risk behavior data of the risk employees in the employee database, or risk behavior data with inaccurate data is stored, the risk behavior data of the risk employees of at least two employee types in the employee database can be screened. The risk behavior data of risk employees of various employee types can be input into a preset nonlinear model, so that the confidence of each risk behavior data is calculated based on the nonlinear model. The confidence level of the risk behavior data is used for indicating the credibility of the risk behavior data, namely whether the risk behavior data is data for recording the risk behaviors of the risk staff. When the confidence coefficient of certain risk behavior data is not in the preset confidence coefficient interval, the risk behavior data is indicated to be invalid risk behavior data, and the risk behavior data is not acquired at the moment to serve as risk behavior sample data. Therefore, the risk behavior data with the confidence in the preset confidence interval can be determined as risk behavior sample data, and further the risk behavior sample data of the risk employees of at least two employee types can be obtained. It should be noted that, when risk behavior data of at least two employee types of risk employees is obtained from the employee database, risk behavior data of risk employees corresponding to the employee types in the employee database may also be obtained based on big data analysis. That is, the obtained risk behavior sample data of at least two employee types of risk employees may include multiple risk behavior data corresponding to different types of risk employees. That is, the obtained risk behavior sample data of at least two types of risk employees is used for describing multiple risk behaviors corresponding to the at least two types of risk employees, and one risk behavior sample data is used for describing one risk behavior sample data. The employee types may be determined based on an application scenario of an actual employee management system (e.g., in an employee management system of a financial institution, the employee types may be a financial employee, an investment employee, a sales employee, etc., and in an employee management system of a scientific enterprise, the employee types may be a development employee, a design employee, an operation and maintenance employee, etc.), which is not limited herein. Optionally, the nonlinear model includes, but is not limited to GBDT model, ID3, C4.5, and the like, which are not limited herein. The preset confidence interval can be determined according to the nonlinear model of the practical application and the practical application scene, and is not limited herein.

S22, determining type labels corresponding to the employee types based on risk behavior sample data corresponding to the employee types.

In some possible embodiments, after the risk behavior sample data corresponding to each employee type is obtained, a type tag corresponding to each employee type may be determined to mark the employee type of the employee corresponding to each risk behavior sample. The type tag is only used for marking employee types of risk employees, and specific forms include, but are not limited to, one or more of the combination of words, numbers, letters and character strings. Optionally, since each employee type corresponds to multiple risk behavior sample data, a data segment (including but not limited to one or more combinations of text, numbers, letters, and character strings) for describing the employee type or indicating the employee type may be screened from multiple risk behavior sample data corresponding to each employee to determine the employee type corresponding to each employee, and then generating a type tag according to the determined employee type to mark each employee type. The type of employee corresponds to only one type of tag, multiple risk behavior sample data belonging to the same employee type corresponds to one type of tag, the type of tag can be marked in each risk behavior sample data of the same employee type, and the type of tag can also be marked in a data set formed by multiple risk behavior sample data corresponding to the same employee type, and the specific implementation mode is not limited herein.

S23, performing data format processing on the risk behavior sample data of the risk employees of the at least two employee types to obtain risk behavior sample data with preset data formats.

In some possible embodiments, since the data sources and the acquiring manners of the risk behavior data of the staff of each staff type in the staff database are different, or the risk behavior data of the staff of each staff type can be acquired in different manners when the risk behavior data of the staff of each staff type is acquired based on big data analysis, the risk behavior sample data of the risk staff of at least two staff types have different data formats, such as character type data, numerical type data, and the like. In order to construct a risk comparison information set based on the same data standard and ensure the accuracy of the risk comparison information, the risk behavior sample data of the risk employees of the at least two employee types can be processed in a data format to obtain risk behavior sample data with a preset data format. The preset data format may be determined based on an actual application scenario, which is not limited herein. For example, when the number of the numeric data in the all risk behavior sample data is large and the number of the character data is small, the character data in the all risk behavior sample data may be processed in a data format to take the value of the character string of the character data, so as to obtain the value of the character string of each character type data (which may be obtained by summing up ANSI code values), thereby converting the character type risk behavior sample data into the numeric (preset data format) risk behavior sample data.

In some possible embodiments, after the risk performance sample data of the risk staff of the at least two staff types is converted into the numerical risk performance sample data, the numerical difference between the different risk performance sample data may be caused to be large due to the difference of the data sources and the data original formats of the different risk performance sample data (for example, the numerical value of one risk performance sample data is 0.1, and the numerical value of the other risk performance sample data is 100). When the numerical value difference is large, the data influence weight of the risk behavior sample data with a relatively low numerical value is relatively weakened by the risk behavior sample data with a relatively high numerical value, so that the risk comparison information is error. Therefore, the methods of min-max standardization, z-score standardization, data normalization and the like can be adopted to perform data standardization processing on all the numerical risk behavior sample data so that all the numerical risk behavior sample data are in the same order of magnitude, and the comparability and operability of all the numerical risk behavior sample data are further improved, so that risk comparison information is accurately constructed.

S24, constructing a piece of risk comparison information based on the type label corresponding to each employee type and risk behavior sample data with a preset data format so as to obtain a risk comparison information set at least comprising two pieces of risk comparison information.

In some possible embodiments, after risk behavior sample data with a preset data format corresponding to a risk employee of each employee type is obtained, multiple risk behavior sample data with a preset data format corresponding to the risk employee of each employee type may be generated together with a type tag corresponding to the employee type to generate a piece of risk comparison information. The risk comparison information is used for representing the risk behaviors of risk staff of each worker type. Similarly, risk behavior sample data with preset data formats of multiple risk employees of multiple employee types can be obtained to obtain risk comparison information for verifying whether the behavior data of the employees of various employee types are risk behavior data. Specifically, each risk behavior sample data with a preset data format is part of employee data of a risk employee corresponding to the risk behavior sample data. That is, each of the risk performance sample data with the preset data format may be one or more combinations of a certain character string, a certain text and a certain number in the employee data of the risk employee corresponding to the risk performance sample data, which is not limited herein. Wherein, the part of data only represents a certain risk behavior of a certain risk employee, that is, the part of data is only risk behavior sample data corresponding to a certain risk behavior of a certain risk employee. The part of data can be obtained based on modes such as data analysis and data screening, more effective information can be selected from employee data of risk employees with larger data size to serve as a data basis for constructing risk comparison information, so that the data processing amount of the risk comparison information can be reduced, the prediction accuracy of employee risk early warning is improved, and the applicability is higher.

In some possible embodiments, when a piece of risk comparison information is constructed based on a type tag corresponding to a risk employee of an employee type and a plurality of risk behavior sample data with preset data formats, key data segments of each risk behavior sample data may be extracted first, and the key data segments of each risk behavior sample data may represent key features of a risk behavior corresponding to the risk behavior sample data, so that a plurality of key data segments corresponding to a risk employee of an employee type may be obtained. And the plurality of key data segments and the type tag can be arranged and combined to form a piece of risk comparison information, or the plurality of key data segments can be subjected to data compression to reduce the data size of the plurality of key data segments, and then the risk comparison information is produced together with the type tag. Optionally, in order to avoid the situation that the data types of the multiple risk behavior sample data are not uniform, each risk behavior sample data of a risk employee of one employee type may be subjected to data processing, so that each risk behavior sample data is converted into risk behavior text information, and then feature keywords for describing each risk behavior are extracted from the risk behavior sample information based on a semantic recognition technology, so as to obtain multiple feature keywords for describing multiple risk behaviors of the risk employee of one employee type. Therefore, one piece of risk comparison information can be generated based on various characteristic keywords of risk staff of one staff type and type labels corresponding to the staff type, and a risk comparison information set containing a plurality of pieces of risk comparison information is not difficult to obtain. Wherein, one piece of risk comparison information in the risk comparison information set contains a plurality of characteristic keywords or a plurality of risk behavior key data corresponding to a risk employee of an employee type. And a piece of risk comparison information is used for determining whether a piece of behavior data of an employee type is risk behavior data, i.e. a piece of risk comparison information is used for determining whether an employee behavior of an employee type is risk behavior. It should be specifically noted that the implementation manner shown above may be determined based on the actual application scenario, which is not limited herein.

In some possible embodiments, when the behavior data for recording any one of the worker behaviors is acquired, it may be determined whether the behavior data is risk behavior data based on the constructed risk comparison information set. Specifically, if the risk comparison information set is constructed based on key data segments of multiple risk behavior sample data, the behavior data may be compared with each piece of risk comparison information in the risk comparison information set, and when the behavior data includes key data segments in any risk comparison information in the risk comparison information set, the behavior data may be determined to be risk behavior data. If the risk comparison information set is constructed based on characteristic keywords of various risk behavior sample data, the behavior data can be subjected to data processing to be converted into behavior text information, and behavior keywords for describing employee behaviors corresponding to the behavior data are extracted from the behavior text information based on a semantic recognition technology. And when any piece of risk comparison information in the risk comparison information set contains the behavior keywords corresponding to the behavior data, determining that the behavior data is risk behavior data. It should be specifically noted that, the implementation manner of determining whether the behavior data is risk behavior data needs to be determined according to the construction manner of the risk comparison information set, which is not described herein.

In the embodiment of the invention, by acquiring various risk behavior data of risk employees of various employee types, as much risk comparison information as possible can be obtained, so that the risk comparison information set can accurately determine whether the behavior data of the employees of different employee types are risk behavior data, and the accuracy of determining the risk behavior data is improved. In addition, different construction modes can be adopted to construct the risk comparison information set, so that the accuracy and the flexibility of constructing the risk comparison information set are further improved.

And S13, when determining that the behavior data is risk behavior data based on any risk comparison information in the risk comparison information set, determining a type label corresponding to the any risk comparison information, determining the employee type corresponding to the type label as the employee type of the employee corresponding to the behavior data, and determining the employee corresponding to the behavior data as the risk employee.

In some possible embodiments, when determining that the behavior data is risk behavior data based on any risk comparison information in the risk comparison information set, risk comparison information for determining that the behavior data is risk behavior data may be determined as target risk comparison information. The risk comparison information used for determining that the behavior data is risk comparison information of matching of the contained key data segment with the behavior data, or risk comparison information containing a behavior keyword corresponding to the behavior data. Therefore, after the target risk comparison information is determined, the employee type corresponding to the type tag included in the risk comparison information can be determined, and then the employee type corresponding to the risk comparison information can be determined as the employee type of the employee corresponding to the behavior data. At this time, since the behavior data is risk behavior data, it indicates that the employee behavior recorded by the behavior data is risk behavior, so that the employee corresponding to the behavior data can be determined as a risk employee, and further risk early warning can be performed on the determined risk employee.

S14, acquiring risk information of the behavior data, and determining the risk level of the risk staff based on the risk information.

In some possible embodiments, after determining that the behavior data is risk behavior data, a behavior feature value of the behavior data may be determined, and then a risk level of the risk employee may be determined based on the behavior feature value. Specifically, although the behavior data only records one type of employee behavior, the behavior data still has data irrelevant to the employee behavior, so that the behavior data can be subjected to feature selection to obtain feature data relevant to the employee behavior recorded by the behavior data. The feature selection manner may be based on a Filter method based on ordering, a Wrapper method based on evaluation, a sequence selection algorithm, and the like, which are not limited herein. For example, when selecting features based on the ranked Filter method, each feature in the behavior data may be scored based on a metric corresponding to the behavior data (specific metrics may be determined based on different behavior data and actual application scenarios, and are not limited herein). And then selecting the feature ranked at the top according to the score ranking, and further determining the feature ranked at the top as feature data corresponding to the employee behavior data. At this time, feature vectors corresponding to the behavior data may be constructed based on the feature data obtained through feature selection, and behavior feature values of the behavior data may be calculated further based on the feature vectors corresponding to the behavior data. So far, the risk level of the risk staff can be determined based on the corresponding relation between the behavior characteristic value and the risk level in the preset risk level table. For example, when the preset risk level table specifies that the behavior feature value is smaller than the first preset threshold value, the low risk level is corresponding, when the behavior feature value is between the first preset threshold value and the second preset threshold value, the medium risk level is corresponding, and when the behavior feature value is larger than the second preset threshold value, the high risk level is corresponding. At this time, if the behavior characteristic value of the behavior data is greater than a second preset threshold value in the preset risk level table, it may be determined that the risk level of the risk employee is a high risk level. The specific value settings of the first preset threshold and the second preset threshold may be determined based on the data type in the actual employee database and the actual application scenario, which is not limited herein. It should be specifically noted that, in order to distinguish the risk levels of the risk employees in detail, a plurality of threshold intervals may be set in the preset risk level table to determine risk levels of different degrees based on a plurality of smaller threshold intervals, and specific setting manners are not described herein.

In some possible embodiments, in order to reduce the selection error in the feature selection process, after determining that the behavior data is risk behavior data, the behavior data may be subjected to data parsing to convert the behavior data into risk behavior text, that is, the employee behaviors recorded in the behavior data are represented in the form of risk behavior text, and the abstract behavior data may be visually presented in the form of text. So that the risk behavior text may be semantically identified to determine behavior feature words (e.g., override, outgoing mail, etc., without limitation) describing the employee behavior recorded by the behavior data from the risk behavior text. In the process of converting the behavior data into the risk behavior text, the behavior data can be converted into binary data, and then the binary data is converted into the risk behavior text, so that the difficulty of data conversion is reduced. The behavior data can also be directly converted into risk behavior text information based on a data conversion tool, an algorithm and the like, and the specific implementation mode can be determined based on an actual application scene without limitation. In addition, after determining the risk behavior feature words corresponding to the behavior data, the risk behavior feature words may be matched with each risk keyword in a preset risk keyword library. When a risk keyword (which is an aspect description, hereinafter referred to as a target risk keyword) consistent with or semantically similar to the risk behavior feature word exists in the preset risk keyword library, a risk level corresponding to the target risk keyword may be determined as a risk level of the risk employee. After obtaining a plurality of risk behavior feature words based on the behavior data and determining the target risk keywords corresponding to each risk behavior feature word, the highest risk level in the risk levels corresponding to each target risk keyword may be determined as the risk level of the risk staff, or the risk level of the risk staff may be comprehensively calculated based on the weight proportion occupied by each risk level corresponding to each target risk keyword, and the specific implementation manner is not limited herein.

And S15, generating early warning information based on the employee type, the behavior data and the risk level.

In some possible embodiments, the early warning information may be generated after determining the risk employees to perform risk early warning on the risk employees. The employee information, employee type, early warning measures and the like of the risk employee can be arranged and combined to obtain early warning information, or the employee information, employee type, early warning measures and the like are integrated in a data analysis, data recombination and the like mode to obtain the early warning information, and the specific generation mode of the early warning information is not limited. The employee information includes, but is not limited to, name, gender, work age, time of job entry, attendance record, etc., which is not limited herein. It should be noted that, the early warning information may further include information such as risk behaviors, occurrence time of risk behaviors, and sending manner of early warning information of the risk employee, and specifically, related data may be obtained from employee data of the risk employee in the employee database, and early warning information may be generated together with the employee information, employee type, early warning measure, and the like, which is not limited herein.

In some possible embodiments, the early warning measures may be determined based on the risk level of the risk employee. Specifically, it may be determined whether the risk level of the risk employee is greater than a preset risk level. If the risk level of the risk staff is not greater than the preset risk level, the preset early warning measures can be determined to be the early warning measures corresponding to the risk staff. In other words, when the risk level of the risk employee is relatively small, that is, the negative effect caused by the risk behavior of the risk employee is not great, for example, a low risk level is obtained based on behavior data such as incorrect operation of the employee, a small amount of offensive reimbursement of the employee (e.g., 10 yuan), and the like, at this time, unified preset early warning measures (including but not limited to internal notification, simple warning, and the like, and not limited thereto) may be adopted as the early warning measures corresponding to the low risk level to simplify the early warning process. If the risk level of the risk staff is determined to be greater than the preset risk level, the early warning measures corresponding to the risk staff can be determined based on the staff behaviors recorded by the behavior data of the risk staff (namely, the risk behaviors of the risk staff). Optionally, if the risk level of the risk employee is determined based on a preset risk level table, the behavior feature of the risk employee may be determined based on feature data obtained in the process of determining the employee type. If the risk level of the risk employee is determined based on a preset risk keyword library, determining the behavior feature word of the employee behavior obtained in the employee type determining process as the behavior feature of the risk employee. After the behavior characteristics of the risk staff are obtained, the early warning measures corresponding to the risk staff can be determined based on the behavior characteristics, and the early warning measures corresponding to the risk staff can be determined based on the behavior characteristics and staff data associated with the behavior characteristics. For example, when the behavior feature indicates that the risk employee has a illegal reimbursement behavior, information such as reimbursement time, reimbursement amount, reimbursement times, accumulated reimbursement amount and the like can be obtained from employee data of the risk employee, so as to determine corresponding early warning measures (such as handover supervision team and the like) according to main factors such as the illegal reimbursement amount and the like and based on preset early warning measure standards.

S16, sending the early warning information to a risk early warning platform to perform risk early warning on the risk staff.

In some possible embodiments, after the early warning information is sent to the risk early warning information platform, the risk early warning platform may be instructed to display information such as employee type, employee information, risk level, and employee behavior in the early warning information. Optionally, the risk early warning platform may instruct the risk early warning platform to perform risk early warning on the risk staff based on the early warning measures based on information such as staff type, staff information, risk level and the like in the early warning information. For example, the risk early warning platform may be instructed to send the early warning measure to a designated early warning information receiving person based on the employee type in the early warning information (e.g., if the risk employee is a sales employee, the early warning measure is sent to a sales executive). For another example, the risk early-warning platform may instruct the risk early-warning platform to take the timeliness of risk early-warning for the risk staff according to the early-warning measures based on the risk level in the early-warning information (if the risk level is the highest risk level, the risk early-warning platform needs to immediately take measures for the risk staff according to the early-warning measures in the early-warning information). The specific manner of instructing the risk early-warning platform to perform risk early-warning on the risk staff may be determined based on the specific content and the actual application scenario of the early-warning information, which is not limited herein.

In the embodiment of the invention, whether the behavior data of any employee in the employee database is the risk behavior data can be determined based on the risk comparison information set, and then the risk employee and the employee type, employee information, risk level and other information of the risk employee are determined under the condition that the behavior data of any employee is determined to be the risk behavior data. Based on the implementation mode, real-time detection of all behavior data in the employee database can be realized, and risk behavior data and risk employees can be determined in time. Meanwhile, corresponding early warning measures are determined based on various information such as employee types, employee information and risk grades of risk employees, and the risk early warning platform is indicated to perform risk early warning on the risk employees, so that the situation of inaccurate early warning caused by manual operation or subjective factors is avoided, in addition, an early warning mode can be flexibly adopted based on different information contents in early warning information, flexibility of employee risk early warning is improved, and applicability is high.

Referring to fig. 3, fig. 3 is a schematic structural diagram of an employee risk early warning device according to an embodiment of the present invention. The employee risk early warning device provided by the embodiment of the invention comprises:

an acquiring unit 31 for acquiring behavior data for recording any one of the worker behaviors from the employee database;

a judging unit 32 for determining whether the behavior data is risk behavior data based on the risk comparison information set;

A determining unit 33, configured to determine, when determining that the behavior data is risk behavior data based on any risk comparison information in the risk comparison information set, a type tag corresponding to the any risk comparison information, determine an employee type corresponding to the type tag as an employee type of an employee corresponding to the behavior data, and determine an employee corresponding to the behavior data as a risk employee; the determining unit 33 is configured to obtain risk information of the behavior data, and determine a risk level of the risk employee based on the risk information;

and the early warning unit 34 is configured to generate early warning information based on the employee type, the behavior data and the risk level, and send the early warning information to a risk early warning platform to perform risk early warning on the risk employee.

In some possible embodiments, the above-mentioned acquisition unit 31 is further configured to:

the employee risk early warning device further comprises:

The construction unit 35 is further configured to construct a risk comparison information set based on risk behavior sample data of risk employees of the at least two employee types.

With reference to the second aspect, in a possible implementation manner, the determining unit 33 is configured to:

In some possible embodiments, the determining unit 33 is configured to:

The employee risk early warning device comprises:

The processing unit 36 is configured to perform data format processing on risk behavior sample data of risk employees of the at least two employee types, so as to obtain risk behavior sample data with a preset data format;

the construction unit 35 is configured to construct a piece of risk comparison information based on the type tag corresponding to each employee type and risk behavior sample data with a preset data format, so as to obtain a risk comparison information set including at least two pieces of risk comparison information;

In some possible embodiments, the risk information includes a behavior feature value and/or a risk behavior feature word; the above-mentioned determination unit 33 is for:

The processing unit 36 is configured to perform data analysis on the risk behavior data to convert the risk behavior data into risk behavior text;

The processing unit 36 is configured to perform semantic recognition on the risk behavior text to determine a risk behavior feature word of the risk employee from the risk behavior text;

the determining unit 33 is configured to determine a risk level of the risk employee based on the risk behavior feature word and a preset risk keyword library.

In some possible embodiments, the above-mentioned obtaining unit 31 is configured to:

obtaining employee information of the risk employee from the employee database;

The judging unit 32 is configured to determine whether the risk level is greater than a preset risk level;

The determining unit 33 is configured to determine, based on the employee behavior recorded by the behavior data, an early warning measure corresponding to the risk employee when the risk level is greater than a preset risk level, and determine, when the risk level is not greater than the preset risk level, the preset early warning measure as the early warning measure corresponding to the risk employee;

The early warning unit 34 is configured to generate early warning information based on the employee information, the employee type, and the early warning measures, and send the early warning information to a risk early warning platform, so that the risk early warning platform displays the employee information and the employee type of the risk employee, and performs risk early warning on the risk employee based on the early warning measures in the early warning information.

In a specific implementation, the employee risk early warning device may execute the implementation provided in the steps of fig. 1 to 2 through each module and/or unit built in the employee risk early warning device. For example, the above-mentioned obtaining unit 31 may be configured to obtain, from an employee database, the behavior data for recording any one of the employee behaviors, and the like, and specifically, the implementation provided by each step may be referred to, which is not described herein. The determining unit 32 may be configured to determine whether the behavior data is risk behavior data based on the risk comparison information set, and the like, and specifically refer to the implementation manners provided by the foregoing steps, which are not described herein. The determining unit 33 may be configured to determine, when any risk comparison information in the risk comparison information set determines that the behavior data is risk behavior data, an implementation manner of a type tag corresponding to the any risk comparison information, and the like, and specifically refer to an implementation manner provided by each step, which is not described herein. The early warning unit 34 may be configured to generate early warning information based on the employee type, the behavior data and the risk level, and send the early warning information to a risk early warning platform to perform risk early warning on the risk employee, and the implementation manner provided by each step may be referred to specifically, and will not be described herein. The construction unit 35 may be configured to construct an implementation manner such as a risk comparison information set based on risk behavior sample data of risk employees of the at least two employee types, and specifically, the implementation manner provided by each step may be referred to, which is not described herein. The processing unit 36 may be configured to perform data analysis on the risk performance data to convert the risk performance data into a risk performance text, and the implementation manner provided by each step may be referred to herein, and will not be described in detail.

Referring to fig. 4, fig. 4 is a schematic structural diagram of a terminal according to an embodiment of the present invention. As shown in fig. 4, the terminal in this embodiment may include: one or more processors 41 and a memory 42. The processor 41 and the memory 42 are connected by a bus 43. The memory 42 is for storing a computer program comprising program instructions, the processor 41 being for executing the program instructions stored by the memory 42 for performing the following operations:

In some possible embodiments, the above processor 41 is further configured to:

In some possible embodiments, the above-mentioned processor 41 is configured to:

Determining a type label corresponding to each employee type based on risk behavior sample data = corresponding to each employee type; processing the risk behavior sample data of the risk employees of the at least two employee types in a data format to obtain risk behavior sample data with a preset data format;

In some possible embodiments, the risk information includes a behavior feature value and/or a risk behavior feature word; the processor 41 is configured to:

And/or, carrying out data analysis on the risk behavior data to convert the risk behavior data into risk behavior text;

obtaining employee information of the risk employee from the employee database;

determining whether the risk level is greater than a preset risk level;

If the risk level is greater than a preset risk level, determining an early warning measure corresponding to the risk staff based on staff behaviors recorded by the behavior data, and if the risk level is not greater than the preset risk level, determining the preset early warning measure as the early warning measure corresponding to the risk staff;

It should be appreciated that in some possible embodiments, the processor 41 may be a central processing unit (central processing unit, CPU), the processor 41 may also be other general purpose processors, digital signal processors (DIGITAL SIGNAL processors, DSPs), application Specific Integrated Circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 42 may include read only memory and random access memory and provides instructions and data to the processor 41. A portion of memory 42 may also include non-volatile random access memory. For example, the memory 42 may also store information of the device type.

In a specific implementation, the terminal may execute, through each functional module built in the terminal, an implementation manner provided by each step in fig. 1 to 2, and specifically, the implementation manner provided by each step may be referred to, which is not described herein again.

In the embodiment of the invention, whether the behavior data of any employee in the employee database is the risk behavior data can be determined based on the risk comparison information set, and then the risk employee and the employee type, employee information, risk level and other information of the risk employee are determined under the condition that the behavior data of any employee is determined to be the risk behavior data. Based on the implementation mode, real-time detection of all behavior data in the employee database can be realized, and risk behavior data and risk employees can be determined in time. Meanwhile, corresponding early warning measures are determined based on various information such as employee types, employee information and risk grades of risk employees, and the risk early warning platform is indicated to perform risk early warning on the risk employees, so that the situation of inaccurate early warning caused by manual operation or subjective factors is avoided, and the applicability is high.

The embodiment of the present invention further provides a computer readable storage medium, where a computer program is stored and executed by a processor to implement the method provided by each step in fig. 1 to 2, and specifically, the implementation manner provided by each step may be referred to, which is not described herein.

The computer readable storage medium may be the task processing device provided in any one of the foregoing embodiments or an internal storage unit of the terminal, for example, a hard disk or a memory of an electronic device. The computer readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD), or the like, which are provided on the electronic device. The computer readable storage medium may also include a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (randomaccess memory, RAM), or the like. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of the electronic device. The computer-readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

The terms first, second and the like in the claims and in the description and drawings are used for distinguishing between different objects and not for describing a particular sequential order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus. Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments. The term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

The foregoing disclosure is illustrative of the present invention and is not to be construed as limiting the scope of the invention, which is defined by the appended claims.

Claims

1. An employee risk early warning method, comprising:

Acquiring risk behavior sample data of risk staff of at least two staff types from a staff database, wherein one staff type corresponds to a plurality of risk staff, one risk staff corresponds to at least two risk behavior sample data, one risk behavior sample data is used for recording one risk behavior, the staff type is determined based on an application scene of an actual staff management system, the staff type comprises financial staff, investment staff and sales staff in the staff management system of a financial institution, and acquiring the risk behavior sample data of the risk staff of the at least two staff types from the staff database comprises: determining risk employees of at least two employee types from an employee database; inputting risk behavior data of risk staff of various staff types into a preset nonlinear model, and obtaining confidence degrees of various risk behavior data based on the preset nonlinear model, wherein the confidence degrees of the risk behavior data are used for representing the credibility degree of the risk behavior data, and the nonlinear model comprises a GBDT model; determining risk behavior data with confidence degrees within a preset confidence degree interval as risk behavior sample data to obtain risk behavior sample data of risk employees of the at least two employee types;

Determining type labels corresponding to each employee type based on risk behavior sample data corresponding to each employee type, wherein the type labels are used for marking employee types of risk employees and comprise the following steps: screening data segments for describing employee types or indicating the employee types from various risk behavior sample data corresponding to each employee to determine the employee type corresponding to each employee, and marking each employee type according to a type label generated by the determined employee type, wherein one employee type only corresponds to one type label, and various risk behavior sample data belonging to the same employee type corresponds to one type label;

performing data format processing on the risk behavior sample data of the risk employees of the at least two employee types to obtain at least two risk behavior sample data with preset data formats, wherein the risk behavior sample data comprises: performing data format processing on character type data in all risk behavior sample data, converting the character type risk behavior sample data into numerical type risk behavior sample data, and performing data standardization processing on all numerical type risk behavior sample data by adopting a method comprising min-max standardization, z-score standardization and data standardization so that all numerical type risk behavior sample data are in the same order of magnitude;

Constructing a piece of risk comparison information based on the type label corresponding to each employee type and at least two pieces of risk behavior sample data with preset data formats to obtain a risk comparison information set at least comprising two pieces of risk comparison information, wherein the piece of risk comparison information comprises a plurality of characteristic keywords or a plurality of risk behavior key data corresponding to risk employees of one employee type; wherein, one piece of risk comparison information in the risk comparison information set is used for comparing with the behavior data of one employee type to determine whether one behavior data of one employee type is risk behavior data, and the method comprises the following steps: extracting key data segments of each risk behavior sample data, wherein the key data segments of each risk behavior sample data are used for showing key characteristics of risk behaviors corresponding to each risk behavior sample data, obtaining a plurality of key data segments corresponding to risk employees of one employee type, and arranging and combining the plurality of key data segments with the type labels to form a piece of risk comparison information, wherein one piece of risk comparison information in the risk comparison information set comprises a plurality of characteristic keywords or a plurality of risk behavior key data corresponding to risk employees of one employee type, and the one piece of risk comparison information is used for determining whether one piece of behavior data of one employee type is the risk behavior data;

determining whether the behavioral data is risk behavioral data based on a risk comparison information set;

Acquiring risk information of the behavior data, and determining the risk level of the risk staff based on the risk information, wherein the risk information comprises a behavior characteristic value and/or a risk behavior characteristic word; the acquiring the risk information of the behavior data and determining the risk level of the risk staff based on the risk information comprises the following steps: determining a behavior characteristic value of the behavior data, and determining the risk level of the risk employee based on the behavior characteristic value and a preset risk level table; and/or performing data analysis on the risk behavior data to convert the behavior data into risk behavior text; performing semantic recognition on the risk behavior text to determine risk behavior feature words of the risk staff from the risk behavior text; determining the risk level of the risk staff based on the risk behavior feature words and a preset risk keyword library;

2. The method of claim 1, wherein the generating pre-warning information based on the employee type, the behavioral data, and the risk level, and sending the pre-warning information to a risk pre-warning platform to risk pre-warn the risk employee comprises:

obtaining employee information of the risk employee from the employee database;

Determining whether the risk level is greater than a preset risk level;

if the risk level is greater than a preset risk level, determining an early warning measure corresponding to the risk employee based on employee behaviors recorded by the behavior data, and if the risk level is not greater than the preset risk level, determining the preset early warning measure as the early warning measure corresponding to the risk employee;

generating early warning information based on the employee information, the employee type and the early warning measures, and sending the early warning information to a risk early warning platform, so that the risk early warning platform displays the employee information and the employee type of the risk employee, and performing risk early warning on the risk employee based on the early warning measures in the early warning information.

3. An employee risk early warning apparatus for performing the method of claim 1 or 2, the employee risk early warning apparatus comprising:

the determining unit is used for determining a type label corresponding to any risk comparison information when determining that the behavior data is risk behavior data based on any risk comparison information in the risk comparison information set, determining the employee type corresponding to the type label as the employee type of the employee corresponding to the behavior data, and determining the employee corresponding to the behavior data as a risk employee;

4. A terminal comprising a processor and a memory, the processor and the memory being interconnected;

The memory is for storing a computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method of claim 1 or 2.

5. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program, which is executed by a processor to implement the method of claim 1 or 2.