员工离职可视化分析及倾向预测研究

Research on Visualization Analysis and Tendency Prediction of Employee Turnover

本文选择GL互联网公司为研究对象,采集员工近3年的数据信息,运用统计分析软件对员工数据进行可视化分析,同时运用多种机器模型包括支持向量机(SVM)、XGBoost和决策树对员工进行离职预测。可视化结果显示工龄、年龄、职级和职位类别等是员工离职的主要影响因素,离职测试结果显示XGBoost预测结果较其他算法较准确,预测离职率准确性较高,在研究结果的基础上为企业提出建立员工信息系统、完善晋升制度和优化薪酬结构等对策建议,以减少企业人员流失。

GL Internet company is selected as the research object,the data information of employees in recent three years is collected,and the visualization analysis of the data of employees is taken by using statistical analysis software.Meanwhile,several machine learning models,including support vector machine(SVM),XGBoost and decision tree,are used to predict employee turnover.The visualization results show that length of service,age,rank and position category are the main influencing factors of employee turnover.The turnover test results show that the prediction results of XGBoost are more accurate than other algorithms,and the accuracy of predicting turnover rate is higher.Based on the research results,countermeasures such as establishing employee information system,improving promotion system and optimizing salary structure are proposed for enterprises to reduce the turnover of enterprise staff.