A Road Map to Enhance Employability Index and Selection Prediction of Management Students Using K-means Clustering and Binary Regression

This paper suggests a roadmap to improve the employability of management students by using some popular data-mining techniques. K-Means clustering and binary regression are used for the purpose of clustering and predicting employability accuracy respectively. The aim of this paper is to present the innovative application of these techniques in higher education for enhancing the employability of professional students that poses a major concern. Since it has been found in a study by ASSOCHAM that 90% of the management graduates in India are not job ready, hence, the demand for such courses is also sharply declining. Data-mining techniques can be applied to meet such challenges. The paper focuses on two clustering algorithm, hierarchal and K-means to first finalise the appropriate numbers of cluster using SPSS 13 by taking a sample of 300 management students of different colleges and then using binary logistic regression to measure the predictive ability of the selection of students. The study works on two hypotheses are as follows: H1: Clustering enhances the performance of students in placement process; H2: Prediction accuracy increases significantly using binary logistic regression. Both of these hypotheses have been proved using an experimental research design where sample will be studied twice in a gap of 3 months.