Volume 18, No. 5, 2021

The Application of Data Mining and BI to the Study of and Efforts to Enhance Academic Performance


Rupa Khanna Malhotra , Narendra Singh Bohra , Pravin P Patil , Durgaprasad Gangodkar , Dibyahash Bordoloi

Abstract

The higher education sector has expressed worry about the high rate of academic failure among university students. The first year of college has the highest attrition rate because students leave because of their academic performance. There has been much discussion and effort put into attempting to determine what may have led to this disappointing showing. Therefore, there are numerous potential applications for the capacity to forecast a student's achievement in higher education. The suggested approach makes use of data mining methods to determine what factors significantly impact and influence undergraduate students' performance. The academic pattern is then analysed using students' demographic and previous academic achievement data. The information is buried in the school database, waiting to be uncovered by data mining tools. Midterm assessments, end-of-semester examinations, abilities, ethics, grasping capacity, extracurricular activities, and other educational data sets may all be mined for this kind of information. Such extraction is facilitated by data categorization algorithms combined with decision trees, and the resulting analysis may be used to generate semantic rules for foretelling students' final grades. Semantic web technologies like ontologies and semantic rules are used by the system to improve the quality of the lessons and exercises that are given to each individual learner. The suggested method inspires a sense of trust among educators and their pupils. As a result, the system intends to analyse the extracted data and mine educational data to generate graphical and statistical results that can aid in the enhancement of student performance and provide instructors with an overall picture of the student's level of proficiency in their chosen field of study.


Pages: 3064-3070

Keywords: Data mining, ID3, Naive Bayes, the Perceptron learning rule, and student performance.

Full Text