Talent classification using support vector machine technique / Hamidah Jantan, Norazmah Mat Yusof and Mohd Hanapi Abdul Latif

• Main Authors: Jantan, Hamidah, Mat Yusof, Norazmah, Abdul Latif, Mohd Hanapi
• Format: Research Reports
• Language: English
• Published: 2014
• Subjects: Electronic data processing. Maple (Computer file); Data processing; Analysis; Analytical methods used in the solution of physical problems; Electronic Computers. Computer Science; Evolutionary programming (Computer science). Genetic algorithms; Data mining
• Online Access: https://ir.uitm.edu.my/id/eprint/52722/1/52722.pdf; https://ir.uitm.edu.my/id/eprint/52722/

Database or data warehouse is rich with hidden information that can be used to provide intelligent decision using data mining technique. Data mining is a widely used approach for knowledge discovery in machine learning. Besides, classification and prediction are among the popular tasks in machine learning especially for information elicitation. There are many areas adapted this approach such as in finance, medical, marketing, stock, telecommunication, manufacturing, health care, education, customer relationship and etc. However, the used of this approach has not attracted much attention in Human Resource (HR) field. Databases in HR can provide a rich resource for knowledge discovery especially for HR intelligent decision system development. Soft computing technique is used for information processing by employing methods, which are capable to deal with imprecision and uncertainty issues. By implementing soft computing techniques in data mining especially in HR field can enhance the knowledge discovery process for intelligent decision system. Support Vector Machine (SVM) is among the popular learning algorithm for classification in soft computing techniques. Due to that reason, this study attempts to use SVM algorithm on employee’s performance databases for talent classification. The objective of this study is to suggest the potential classification model for talent forecasting throughout some experiments using SVM learning algorithm. In the experimental phase, we use employee’s performance data from selected organization to develop talent classification model which can be used to handle some tasks in talent management. At the end, the aim of this study is to develop a prototype system using proposed classification model for talent forecasting. Managing talent is among the challenges of HR professionals, especially to ensure the right person for the right job at the right time. Besides, identifying existing talent in an organization is among the top talent management challenge. This task requires a lot of managerial decisions, which are sometimes quite uncertain and difficult. HR decisions depend on various factors such as human experience