基于JSP+DM的人力资源管理系统构建与职工离职预测分析  

Construction of Human Resource Management System and Analysis of Employee Turnover Prediction Based on JSP+DM

作　　者：梁敬 LIANG Jing(Shunyi Hospital of Beijing Traditional Chinese Medicine Hospital,Beijing 101300,China)

机构地区：[1]北京中医医院顺义医院,北京101300

出　　处：《微型电脑应用》2024年第12期308-311,共4页Microcomputer Applications

摘　　要：为了提升大型公立医院人力资源管理效率,方便医院管理者对自身水平进行精准评估,并对可能存在的人事风险进行提前预判。基于Java服务器页面(JSP)和数据挖掘(DM)技术,构建大型公立医院人力资源管理系统。系统可根据每种功能特点选取对应的数据挖掘模型进行分析、预测和管理。针对医院职工离职问题,分别利用先验高斯分布朴素贝叶斯算法和CART决策树算法构建员工离职预测模型,并采用交叉验证法对样本数据进行处理,从而完成员工离职预测分析。结果表明:贝叶斯算法的错误率、精度、查准率和查全率分别为18.17%、81.83%、59.31%和81.85%,而CART决策树算法的错误率、精度、查准率和查全率分别为2.03%、97.97%、94.53%和97.34%,CART决策树算法相比贝叶斯算法更适用于大型公立医院系统的职工离职预测。In order to improve the efficiency of human resource management in large public hospitals,facilitate hospital managers to accurately evaluate their own level,and predict possible personnel risks in advance,a large public hospital human resource management system is built based on Java server pages(JSP)and data mining(DM)technology.The system can select the corresponding data mining model according to each functional feature for analysis,prediction and management.To solve the problem of employee turnover in hospitals,the naive Bayesian algorithm with a priori Gaussian distribution and CART decision tree algorithm are used to build employee turnover prediction models respectively,and the cross validation method is used to process the sample data,so as to complete the prediction analysis of employee turnover.The results show that the error rate,accuracy,precision and recall of Bayesian algorithm are 18.17%,81.83%,59.31%and 81.85%,respectively,while the error rate,accuracy,precision and recall of CART decision tree algorithm are 2.03%,97.97%,94.53%and 97.34%,respectively.Compared with Bayesian algorithm,CART decision tree algorithm is more suitable for employee turnover prediction in large public hospital systems.

关 键 词：大型公立医院 人力资源管理系统 Java服务器页面 数据挖掘 离职预测 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]