Higher education student dropout prediction and analysis through educational data mining

Abstract

Listen

Translate article

In past years the number of student dropout from the educational institute is increasing rapidly. The high rate of student's dropout in a registered course has been a major threat to many educational institutions or universities. The student enters the institution with lots of dreams and expectations. When their expectations are not fulfilling or certain factors like demographics will effect and makes them drop from their registered program. It is a major threat to all educational institution. The various technique of the dimensionality reduction, which includes feature selection and feature extraction. Feature selection is step by step procedure that is used to select the right attribute from a given attribute sets. For the feature extraction process, it involves the transformation of higher dimensions' data in corresponding lower dimensions. Feature selection consists of factors like Academics, demographical factors, psychological factors, health issues, teacher's opinion, student behavior. In this paper, we introduce a methodology to predict the student dropout using Naive-Bayes Classification Algorithm in R language. And also examine the reason for student drop out at an early state and predict whether the student will drop or not. There are many factors that affect a student to commit dropout as we mentioned above. Early dropout prediction helps the organization to retain the students from the respective academic program.