Analyzing the Application of SMOTE on Machine Learning Classifiers

Data mining functionalities are used in almost every sector including education. Due to continuous evolving information technology and decreasing cost of hardware, every educational institute has an urge to keep the students as much as possible data, in their databases to mine unknown insights. These unknown insights make institutes more proactive and counter reactive and helps them to induce more quality to their teaching and learning processes. In this paper, supervised machine learning techniques are used to predict the student performance using classification. Rapid miner 9.6, leading data science platform, is used for comparative analysis of predictions made by k nearest neighbor, naïve bayes, chi square automatic interaction detection decision tree, and random forest using synthetic minority over-sampling technique.