PREDICTING UNDERGRADUATES’ EMPLOYABILITY USING NAÏVE BAYES CLASSIFICATION METHOD

Abstract

Graduate employability is a crucial concern to all countries in the world. Many graduates are looking for job opportunities despite their qualifications and skills due to various factors. Therefore, predicting the employability of undergraduates considering their performances has become a timely requirement as it can help to address unemployment in great success since it provides students an opportunity to groom. This research analyses demographic and educational factors affecting an individual’s employability and develops a model to predict an undergraduates’ capability of achieving career targets beforehand. Data mining methods are often implemented nowadays for analyzing available data and extracting information and knowledge to support decision-making in various domains. In this study, the Naive Bayes classification algorithm is used to predict the possibility of undergraduates’ employability. This paper presents the results of potential employability of undergraduates using demographic and educational information collected via an online survey conducted through social media platforms. Apart from demographic factors, undergraduates’ lecture participation rate, computer literacy, English literacy, intern participation, club memberships, and many more attributes were used to evaluate their competencies. Overall, individuals actively involved with academic activities, including extra-curricular activities, were more likely to be predicted as employed. Therefore, using the naïve Bayes classification algorithm, the most influencing factors that can be used to predict undergraduates’ employability are presented by using a trained model with graduate data with an accuracy of 75.89 %