Applicability of Naïve Bayes Model for Automatic Resume Classification

Abstract

Resume selection and classification is a very important function of Human Resource Department of every institution. Due to increased use of technology and online job application, this department receives large volumes of resumes which has made resume selection and classification a complex process in terms of information processing, time taken and transparency in the selection process. In this research, a machine learning model is proposed to assist resume selection and classification. Naïve Bayes model was developed to select and classify resumes. The predictive accuracy attained will be recorded and compared to predictive accuracy of homogeneous Ensemble classifier model developed by using different data sets. Naïve Bayes classifier models obtained from different data sets was used as base classifiers to develop Ensemble Naïve Bayes Classifier. It was observed that the new model produced a better predictive accuracy compared to the original Naïve Bayes Classifier model. The original Naïve Bayes classifier model gave an average predictive accuracy of 89.8148% while Ensemble Naïve Bayes Classifier model attained an overall accuracy of 94.4444%.