Welcome to the InfoSci Platform
Identifying Non-Performing Students in Higher Educational Institutions Using Data Mining Techniques

Identifying Non-Performing Students in Higher Educational Institutions Using Data Mining Techniques

Deepti Aggarwal, Sonu Mittal, Vikram Bali
Copyright: © 2021 |Volume: 12 |Issue: 1 |Pages: 17
ISSN: 1947-8186|EISSN: 1947-8194|EISBN13: 9781799861508|DOI: 10.4018/IJISMD.2021010105
Cite Article Cite Article

MLA

Aggarwal, Deepti, et al. "Identifying Non-Performing Students in Higher Educational Institutions Using Data Mining Techniques." IJISMD vol.12, no.1 2021: pp.94-110. http://doi.org/10.4018/IJISMD.2021010105

APA

Aggarwal, D., Mittal, S., & Bali, V. (2021). Identifying Non-Performing Students in Higher Educational Institutions Using Data Mining Techniques. International Journal of Information System Modeling and Design (IJISMD), 12(1), 94-110. http://doi.org/10.4018/IJISMD.2021010105

Chicago

Aggarwal, Deepti, Sonu Mittal, and Vikram Bali. "Identifying Non-Performing Students in Higher Educational Institutions Using Data Mining Techniques," International Journal of Information System Modeling and Design (IJISMD) 12, no.1: 94-110. http://doi.org/10.4018/IJISMD.2021010105

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The educational institutes are focusing on improving the performance of students by using several data mining techniques. Since there is an increase in the number of drop out students every year, if we are able to predict whether a student will complete the course or not, it is possible to take some preventive actions beforehand. The primary data set used for modelling has been taken from a reputed technical institute of Uttar Pradesh which consists of data of 6,807 students containing 20 academic and non-academic attributes. The most relevant attributes are extracted using CorrelationAttributeEval (in WEKA) technique using Ranker search method which ranks the attributes as per their evaluation. Synthetic minority oversampling technique (SMOTE) filter is applied to deal with the skewed data set. The models are built from eight classifiers that are analysed for predicting the most appropriate model to classify whether a student will complete the course or withdraw his/her admission.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.