ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

A Review on Prediction of Academic Performance of Students at-Risk Using Data Mining Techniques

Journal: Journal on Today's Ideas - Tomorrow's Technologies (Vol.5, No. 1)

Publication Date:

Authors : ; ;

Page : 30-39

Keywords : Data Mining (DM); Educational data mining (EDM); Education System;

Source : Downloadexternal Find it from : Google Scholarexternal


Educational data mining is the procedure of converting raw data collected from educational databases into some useful information. It can be helpful in designing and answering research questions like performance prediction of students in academics, factors that affect the students' performance, help the teachers in understanding the problems faced by the students to understand the course content and complexity of the subject taken so that the teachers can take timely action to control the dropout rate. This also includes improving the teaching-learning process so that the interventions can be taken at the right time to improve the performance of the student. This paper is the review of the research work done in the field of educational data mining for the prediction of students' performance. The factors that influence the performance of the students i.e. the type of classrooms they attend such as traditional or online, socio-economic, educational background of the family, attitude toward studies and challenges faced by the students during course progress. These factors lead to the categorization of the students into three groups “Low-Risk”: who have a High probability of succeeding, “Medium-Risk”: who may succeed in their examination, “High-Risk”: who have a High probability of failing or drop-out. It elaborates the different ways to improve the teaching-learning process by providing the students personal assistance, notes, class-assignments and special class tests. The most efficient techniques that are used in educational data mining are also reviewed such as; classification, regression, clustering, and prediction.

Last modified: 2018-05-15 20:47:14