| 摘 要: 人力资源管理在决策方式上逐渐智能化。为了辅助人力资源进行员工晋升决策，提高管理者晋升决策的公平性和有效性，提出基于集成学习的极端梯度提升(XGBoost)模型对数据分布不平衡的员工晋升数据进行预测分析。以Kaggle(数据科学竞赛平台)的员工晋升数据集为对象进行预处理，建立基于XGBoost算法的员工晋升预测模型，结合准确率、F1值和AUC值这几个评价指标，与其他算法模型进行比较分析。实验结果表明，相比逻辑回归(LR)、支持向量机(SVM)、人工神经网络(ANN)、多层感知机(MLP)模型，XGBoost模型的三项评价指标更具优势，应用于员工晋升预测，效果更好。 |
| 关键词: 人力资源；员工晋升预测；机器学习；XGBoost |
|
中图分类号: TP319
文献标识码: A
|
| 基金项目: 浙江省重点研发计划项目(2022C01207). |
|
| Unbalanced Employee Promotion Prediction based on XGBoost |
|
HUANG Jing1, ZHENG Huihui2
|
( 1.School of Information Science and Engineering, Zhejiang Sci -Tech University, Hangzhou 310018, China;
2.School of Computer Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China)
syhj_sy@163.com; 472596438@qq.com
|
| Abstract: Human resource management is becoming more intelligent in decision-making. In order to assist human resources in making employee promotion decisions and improve the fairness and the effectiveness of managers' promotion decisions, this paper proposes an XGBoost model based on integrated learning to predict and analyze employee promotion data with unbalanced data distribution. Taking the employee promotion dataset of Kaggle data science competition platform as the object for preprocessing, employee promotion prediction model is established based on XGBoost algorithm. This model is then compared with other algorithm models on the evaluation indicators of accuracy, F1 value and AUC value. The experimental results show that XGBoost model has more advantages in three evaluation indicators than LR (Logistic Regression), SVM (Support Vector Machine), ANN (Artificial Neural Network) and MLP (Multilayer Perceptron) models, and it has better effect when applied to employee promotion prediction. |
| Keywords: human resources; employee promotion prediction; machine learning; XGBoost |
