A Comparative Study Using Linear Model, Decision Tree and Random Forest for Classification of Student’s Self Satisfaction with Family Aspect Parameters of Students

Abstract

This paper is outcome of an experimental work carried out over educational data where linear model, decision tree and random forest have been deployed over own educational dataset to get knowledge for academic analytics, which otherwise is invisible. The linear model, decision tree and random forest models have been compared and it is observed that the random forest approach is more suitable and gives accurate results. Analytics were carried out with R software package.