Dynamics of business and opportunities for employability using predictive modeling tools of Machine Learning and R programming

Employability is a major concern in India along with poverty, economy, education, pollution, and population growth. To combat the concern, each higher education institution in India is striving hard to comply with the nuances of the market. To enhance the employability of job seekers, it is essential to address certain factors that contribute to their increased employability. This study has identified twelve such variables from the extant literature. Thereafter, the data has been collected from Institutions preferably imparting professional education on the twelve identified variables and the dependent variable, employability. Logistic Regression, Decision Tree, Random Forest, and Naïve Bayes as classification techniques of data mining have been adopted in this study to predict the employability of job seekers. These classification techniques were used and the highest accuracy of 0.97 with Area under ROC was 0.99 received using Random Forest. The accuracy levels and AUC of all the classification techniques are very high, it means these twelve variables have substantial role in classification of the dependent variable, employability. The top seven variables Domain Knowledge, Problem Solving, Computer Skills, General Awareness, Teamwork, Analytical Thinking and Creativity play important roles in classification of employability. Therefore, the management institutes should focus on these variables to enhance the employability of the students. The future scope of this study can be used for other professional programs like MCA, Engineering, Diploma etc.