Academic performances and campus placements decides the student’s carrier and reputation of the educational institutions. Today various methodologies are available to predict or forecast the students’ performance in academic and placement interviews. In the educational data mining (EDM), apart from data mining techniques, Machine Learning (ML) techniques needed to achieve high standard of prediction results to enhance the quality of the education, institution and student performance. An ensemble model from ML helps to combine different models and provides improved results. In this work, Multilayer Perception (MLP), Naïve Bayes tree (NBT), Logistic model tree (LMT) and J48 classifiers are utilized in ensemble voting process to predict the students’ placement possibility. In the first ensemble voting model, BayesNet and J48 classifiers are combined using its mean value of the probability with combination rule and the second ensemble voting model, MLP and LMT are combined using its mean value of the probability, obtained attributed and instances are passed to NBTree for further process. The MLP and LMT based ensemble voting model produced better accuracy (92.8%) than first model (91%) with least error prediction rate.

Campus Placement, Ensemble Voting, Logistic Model Tree, NBTree