The academic institutions are focusing more on improving the performance of students using various data mining techniques. Prediction models are designed to predict the performance of students at a very early stage so that preventive measures can be taken beforehand. Various parameters (academic as well as non-academic) are considered to predict the student performance using different classifiers. Normally, academic parameters are given more weightage in predicting the academic performance of a student. This paper compares the two models: one built using academic parameters only and another using both academic and non-academic (demographic) parameters. The primary data set of students has been taken from a technical college in India, which consists of data of 6,807 students containing attributes. Synthetic minority oversampling technique filter is applied to deal with the skewed data set. The models are built using eight classification algorithms that are then compared to find the parameters that help to give the most appropriate model to classify a student based on his performance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.