基于机器学习的人力薪资管理及绩效优化体系研究

Research on human salary management and performance optimization system under machine learning

针对当前研究对于人力薪资管理及绩效优化并不理想的现状,构建了一种基于机器学习改进薪资和绩效的模型。基于XGBoost和GBDT算法研究了主观与客观因素对薪资的影响,通过优化对其模型准确率进行RMSE误差分析,得出XGBoost算法模型的RMSE22.36,GBDT算法模型的RMSE为39.85;采用层次分析法对绩效各指标权重进行了研究评价,实现了绩效指标的科学设定以及机器学习框架下数据监测的动态调整,验证了该模型对于人力薪资管理及绩效优化体系的有效性。

Based on the fact that the current research is not ideal for human salary management and performance optimization,a model which introduces machine learning is constructed to improve salary and performance.Based on XGBoost and GBDT algorithm,the influence of subjective and objective factors on salary is studied.RMSE error analysis is carried out on the model accuracy rate through optimization,and the accuracy rate of XGBoost algorithm model is RMSE22.36 and that of GBDT algorithm model is 39.85.The analytic hierarchy process is used to study and evaluate the weight of each performance indicator,which realizes the scientific setting of performance indicators and the dynamic adjustment of data monitoring under the framework of machine learning,and verifies the effectiveness of this model for the research of human salary management and performance optimization system.