A Comparative Study on Machine Learning Algorithms for Predicting the Placement Information of Under Graduate Students

As Machine Learning (ML) algorithms are becoming popular to solve challenging and interesting real world prediction problems around us, the interest level of student community has been increased in learning the principles of ML and its different algorithms. This includes by implementing the commonly known machine learning algorithms and tests them by solving simple prediction problems around the student community present in educational system. In this line, this paper proposes to solve the student placement prediction problem using linear regression model, K-neighbor regression model, decision tree regression model, XGBoost regression model, gradient boost regression model, light GBM regression model and random tree classifier model. This work is carried out in two phases. The Phase 1 is done on a simple data set and the Phase 2 is done with an extended data set with added additional features about the students. This research work presents the comparative performance analysis of these seven models by implementing them with these two data sets. The performance measurements considered in this study are prediction accuracy and the root mean square error (RMSE).