CN-121998149-A - Service risk early warning method and device and computer readable storage medium

CN121998149ACN 121998149 ACN121998149 ACN 121998149ACN-121998149-A

Abstract

The invention provides a business risk early warning method, a business risk early warning device and a computer readable storage medium, wherein the method comprises the steps of acquiring employee basic information of a target employee, learning behavior information and learning result information of the target employee in a target business knowledge point type; the basic information of staff, the learning behavior information of the target staff in the target business knowledge point type and the learning result information are input into a trained extreme gradient lifting tree XGBOOST model to obtain a performance level predicted value of the target staff in the target business knowledge point type, and whether the target staff has business risks is judged according to the performance level predicted value of the target staff in the target business knowledge point type. The method, the device and the computer readable storage medium can solve the problem that the prior art lacks a business risk early warning method based on business knowledge learning, so that the business risk caused by insufficient business knowledge learning of staff cannot be timely and effectively found.

Inventors

YE YONGFEI
ZOU YUJIE
WAN XING
ZHOU TIAN
YAO PEIJUN
GONG KE

Assignees

中国联合网络通信集团有限公司

Dates

Publication Date: 20260508
Application Date: 20241106

Claims (10)

1. A business risk early warning method, the method comprising: Acquiring employee basic information of a target employee, learning behavior information and learning result information of the target employee in a target service knowledge point type; Inputting the basic employee information, the learning behavior information of the target employee in the target business knowledge point type and the learning result information into a trained extreme gradient promotion tree XGBOOST model to obtain a performance level prediction value of the target employee in the target business knowledge point type, wherein the trained XGBOOST model is obtained based on employee basic information of a plurality of employees, historical learning behavior information in the target business knowledge point type, historical learning result information and historical performance level training; And judging whether the target staff has business risks or not according to the performance level predicted value of the target staff at the target business knowledge point type.
2. The method of claim 1, wherein the obtaining employee base information of the target employee and learning behavior information and learning result information of the target employee in the target business knowledge point type is preceded by the method further comprising: Collecting unclassified learning content data and topic content data of the target staff in a preset learning platform; classifying the unclassified learning content data and the topic content data through a trained textCNN model to identify the learning content data and the topic content data under different service knowledge point types; And determining learning behavior information and learning result information of the target staff in each business knowledge point type according to the learning content data and the topic content data of the different business knowledge point types, wherein the different business knowledge point types comprise the target business knowledge point type.
3. The method of claim 2, wherein before classifying the unclassified learning content data and the topic content data by the trained textCNN model to identify learning content data and topic content data under different service knowledge point types, the method further comprises: Acquiring learning content data and topic content data of marked business knowledge point types recorded in the learning platform; and constructing and training textCNN a model according to the learning content data and the topic content data of the marked service knowledge point type to obtain a trained textCNN model.
4. The method of claim 1, wherein the learning behavior information includes a learning start time, a number of words of the article, a stay time of the article, a start time of the answer, an answer time, a number of retries of the answer, and an average stay time of the number of words of the article, and the learning result information includes an answer score.
5. The method of claim 1, wherein the inputting the employee base information, learning behavior information of the target employee in a target business knowledge point type, and learning outcome information into a trained extreme gradient boost tree XGBOOST model results in a predicted value of a performance level of the target employee in the target business knowledge point type, the method further comprising: Taking employee basic information of the plurality of employees, historical learning behavior information in the target business knowledge point type and historical learning result information as a characteristic data set; taking the historical performance level of the plurality of staff in the target business knowledge point type as a target variable corresponding to the characteristic data set; And constructing and training the XGBOOST model according to the characteristic dataset and the corresponding target variable to obtain the trained XGBOOST model.
6. The method according to claim 1, wherein the performance level is equal to a best, a middle or a poor, and the determining whether the target employee has a business risk according to the performance level prediction value of the target employee at the target business knowledge point type specifically includes: And if the performance level predicted value of the target employee at the target business knowledge point type is poor, judging that the target employee has business risk.
7. The method of claim 6, wherein if the target employee is at business risk, the method further comprises: and taking relevant measures for coaching the target staff.
8. A business risk early warning device, comprising: the first acquisition module is used for acquiring staff basic information of a target staff, learning behavior information and learning result information of the target staff in a target service knowledge point type; The prediction module is connected with the first acquisition module and is used for inputting the basic information of the staff, the learning behavior information of the target staff in the target business knowledge point type and the learning result information into a trained extreme gradient lifting tree XGBOOST model to obtain a performance level prediction value of the target staff in the target business knowledge point type, wherein the trained XGBOOST model is obtained based on staff basic information of a plurality of staff, the history learning behavior information, the history learning result information and the history performance level training in the target business knowledge point type; and the early warning module is connected with the prediction module and is used for judging whether the target staff has business risk according to the performance level predicted value of the target staff at the target business knowledge point type.
9. A business risk early warning device comprising a memory and a processor, the memory having a computer program stored therein, the processor being arranged to run the computer program to implement the business risk early warning method of any one of claims 1 to 7.
10. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, which when executed by a processor, implements the business risk early warning method according to any one of claims 1-7.

Description

Service risk early warning method and device and computer readable storage medium Technical Field The present invention relates to the field of data processing technologies, and in particular, to a business risk early warning method, a business risk early warning device, and a computer readable storage medium. Background Along with the importance of enterprises on continuous learning of employees, more and more enterprises start to build an internal learning platform so as to improve the professional skills and business level of the employees. However, most of learning platforms are only used for checking whether the staff normally completes the learning task, and the learning process and the learning result of the staff are not effectively analyzed. In the business of enterprises, a great amount of knowledge in vertical fields and artificially defined conceptual knowledge exist, and if staff lacks learning of the knowledge, the staff has great business risks, such as customer loss, data security, low-quality risks and the like, when carrying out marketing or production activities. However, the prior art lacks a business risk early warning method based on business knowledge learning, so that the problem of business risk caused by insufficient business knowledge learning of staff cannot be effectively found in time. Disclosure of Invention The technical problem to be solved by the invention is to provide a business risk early warning method, a device and a computer readable storage medium aiming at the defects in the prior art, so as to solve the problem that the prior art lacks a business risk early warning method based on business knowledge learning, and cannot timely and effectively discover the business risk caused by insufficient business knowledge learning of staff. In a first aspect, the present invention provides a business risk early warning method, including: Acquiring employee basic information of a target employee, learning behavior information and learning result information of the target employee in a target service knowledge point type; Inputting the basic employee information, the learning behavior information of the target employee in the target business knowledge point type and the learning result information into a trained extreme gradient promotion tree XGBOOST model to obtain a performance level prediction value of the target employee in the target business knowledge point type, wherein the trained XGBOOST model is obtained based on employee basic information of a plurality of employees, historical learning behavior information in the target business knowledge point type, historical learning result information and historical performance level training; And judging whether the target staff has business risks or not according to the performance level predicted value of the target staff at the target business knowledge point type. Further, before the staff basic information of the target staff and the learning behavior information and the learning result information of the target staff in the target business knowledge point type are obtained, the method further includes: Collecting unclassified learning content data and topic content data of the target staff in a preset learning platform; classifying the unclassified learning content data and the topic content data through a trained textCNN model to identify the learning content data and the topic content data under different service knowledge point types; And determining learning behavior information and learning result information of the target staff in each business knowledge point type according to the learning content data and the topic content data of the different business knowledge point types, wherein the different business knowledge point types comprise the target business knowledge point type. Further, before classifying the unclassified learning content data and the topic content data by the trained textCNN model to identify the learning content data and the topic content data under different service knowledge point types, the method further includes: Acquiring learning content data and topic content data of marked business knowledge point types recorded in the learning platform; and constructing and training textCNN a model according to the learning content data and the topic content data of the marked service knowledge point type to obtain a trained textCNN model. Further, the learning behavior information comprises learning starting time, article word number, article stay time, answer starting time, answer retry times and article word number average stay time, and the learning result information comprises answer achievements. Further, the step of inputting the basic employee information, learning behavior information of the target employee in the target service knowledge point type, and learning result information into a trained extreme gradient promotion tree XGBOOST model to obtain a performance level predicted value of the