DBGE: Employee Turnover Prediction Based on Dynamic Bipartite Graph Embedding

Xinjun Cai,Wu Xie,Baohua Qiang,Liang Zhao,Feiyi Liu,Ziwei Jin,Jiaxing Shang

doi:10.1109/access.2020.2965544

Abstract

The issue of employee turnover is always critical for companies, and accurate predictions can help them prepare in time. Most past studies on employee turnover have focused on analyzing impact factors or using simple network centrality measures. In this paper, we study the problem from a completely new perspective by modeling users’ historical job records as a dynamic bipartite graph. Specifically, we propose a bipartite graph embedding method with temporal information called dynamic bipartite graph embedding (DBGE) to learn the vector representation of employees and companies. Our approach not only considers the relations between employees and companies but also incorporates temporal information embedded in consecutive work records. We first define the Horary Random Walk on a bipartite graph to generate a sequence for each vertex in chronological order. Then, we employ the skip-gram model to obtain a temporal low-dimensional vector representation for each vertex and apply machine learning methods to predict employee turnover behavior by combining embedded features with employees’ basic information. Experiments on a real-world dataset collected from one of China’s largest online professional social networks show that the features learned through DBGE can significantly improve turnover prediction performance. Moreover, experiments on public Amazon and Taobao datasets show that our approach achieves better performance in the link prediction and visualization task than other graph embedding methods that do not consider temporal information.

Highlights

Employee turnover problems have been widely considered because the resignation of talented employees may reduce a company’s competitiveness
In order to solve the dynamic bipartite graph embedding problem, we extend the random walk method and propose a biased random walk method called Horary Random Walk, which takes the chronological order of the edges into consideration when conducting random walks
DYNAMIC BIPARTITE GRAPH EMBEDDING we describe the framework of our dynamic bipartite graph embedding (DBGE) approach

Summary

INTRODUCTION

Employee turnover problems have been widely considered because the resignation of talented employees may reduce a company’s competitiveness. In order to fully utilize the temporal information and solve the employee turnover prediction problem more effectively, in this paper we propose DBGE, a new Dynamic Bipartite Graph Embedding-based approach. By embedding the employee’s historical work records with the dynamic bipartite graph, we can effectively learn the network-based features for the employee and company vertices. We combine these learned features with basic employee and company features, and feed them into different classic machine learning models to solve the employee turnover prediction problem. We propose a new dynamic bipartite graph embedding method and define the Horary Random Walk to select vertices in chronological order to learn low-dimensional vertex features with temporal information.

10: Append u to walk

EXPERIMENT

CONCLUSION