Predicting students' performance using mutli-criteria classification: A case study

Document Type

Conference Proceeding

Source of Publication

Proceedings of the 28th International Business Information Management Association Conference - Vision 2020: Innovation Management, Development Sustainability, and Competitive Economic Growth

Publication Date

1-1-2016

Abstract

One of the educational data mining goals is to predict students' performance, and analyzing their behavior. Several studies have been conducted to make use of different classifiers to reach this goal. In this research we describe our experience of applying multi-criteria decision aid (MCDA) in the educational data mining domain. We are contributing to this field by studying data analytic techniques applied to real-case studies to predict students' performance according to their past academic experience. Hence, the aim of this research is to utilize MCDA in the education domain. The classification tool that is used to predict students' performance is based several criteria such as: age, school, address, family size, evaluation in previous grades, and activities. Based on the data used used in this paper, we found that some criteria are more influential than others in predicting students' performance. To check the performance, our proposed method was compared with a decision tree classifier, and a comparative and analytical study is conducted on well-known students' data.

ISBN

9780986041983

Publisher

International Business Information Management Association, IBIMA

First Page

1590

Last Page

1599

Disciplines

Computer Sciences

Keywords

Decision tree, Multi-criteria selection, Pre-processing, Students' assessment, Students' performance

Scopus ID

85013884565

Indexed in Scopus

yes

Open Access

no

This document is currently not available here.

Share

COinS