Employee profile and labor turnover in outsourcing companies: A data mining approach

Abstract Data mining techniques can be applied to search for hidden information in large volumes of data. In human resources management, data mining is useful for identifying the reasons behind employee turnover and behavior. This knowledge makes it possible to identify employee profiles and helps improve personnel selection processes, which are appropriate means to reduce company turnover rates. In this article, we analyze the situation of a human resources outsourcing company and apply data mining techniques to classify labor turnover in low-skilled employees. We follow the methodology CRISP-DM to build and evaluate different classification models and discover a list of relevant characteristics of employee profiles prone to turnover. Furthermore, we compare the results of applied techniques to assess performance and suitability to identify factors associated with turnover and generate undesirable employee profiles. The results show that Age, Salary, Location, and Work Experience in Time and Area are key factors that help classify turnover and, therefore, can be used to suggest personnel selection policies to the company. The results obtained in this article may serve as a reference framework for companies that hire low-skilled employees, particularly those that provide human resources outsourcing services, so they can collect and analyze employee data and identify profiles prone to turnover. The significance of this work is that results: i) are presented in the context of a real human resources outsourcing company and ii) are obtained from the analysis of low-skilled employee data available in such a company, which are aspects scarcely explored in related research. A limitation of this research was the partial absence of specific socio-demographic data in the available data set and of variables related to organizational climate and culture.

Bibliographic Details

Main Authors:	Márquez-Hermosillo,Abigail, Rodríguez,Luis Felipe, Salazar-Lugo,Guillermo, Borrego,Gilberto
Format:	Digital revista
Language:	English
Published:	Universidad Nacional Autónoma de México, Facultad de Ingeniería 2023
Online Access:	http://www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1405-77432023000400007
Tags:	Add Tag No Tags, Be the first to tag this record!