Educational data mining for predicting students' academic performance using machine learning algorithms. (2021)

• Record Type: Journal Article
• Authors: Dabhade, Pranav; Agarwal, Ravina; Alameen, K.P.; Fathima, A.T.; Sridharan, R.; Gopakumar, G.
• Is Part Of: Materials today. Volume 47:Part 15(2021)
• Page Start: 5260
• Page End: 5267
• Publication Date: 2021
• Subjects: Educational data mining -- Regression -- Academic performance -- Prediction -- Support vector regression; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.05.646
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: Educational data mining has gained impressive attention in recent years. The primary focus of educational institutions is to provide quality education for students to enhance academic performance. The performance of students depends on several aspects, i.e., personal, academic, and behavioural features. The present study deals with predicting students' academic performance in a technical institution in India. A dataset was obtained using a questionnaire-based survey and the academic section of the chosen institution. Data-pre-processing and factor analysis have been performed on the obtained dataset to remove the anomalies in the data, reduce the dimensionality of data and obtain the most correlated feature. The Python 3 tool is used for the comparison of machine learning algorithms. The support vector regression_linear algorithm provided superior prediction.