Supervised and Unsupervised Learning in Data Mining for Employment Prediction of Fresh Graduate Students

Abstract

Data mining techniques are widely used in engineering, medicine, industry, agriculture and even used in education to predict a future situation. In this paper, the used of data mining techniques applied in features selection and determine the best model that can be used to predict the employment status of fresh graduate Public Institutions either employed or unemployed, six months after graduation. In CRISP-DM methodology, six phases were adopted. The algorithm in supervised and unsupervised learning; K-Nearest Neighbor, Naive Bayes, Decision Tree, Neural Network, Logistic Regression and Support Vector Machines were compared using the training data set from Tracer Study to determine the highest accuracy in turn is used as a predictive model. Rapid Miner as a data mining tool was used for data analysis algorithm