(Publisher of Peer Reviewed Open Access Journals)

International Journal of Advanced Computer Research (IJACR)

ISSN (Print):2249-7277    ISSN (Online):2277-7970
Volume-8 Issue-36 May-2018
Full-Text PDF
DOI:10.19101/IJACR.2018.836003
Paper Title : A proposed academic advisor model based on data mining classification techniques
Author Name : Mohamed Hegazy Mohamed and Hoda Mohamed Waguih
Abstract :

University and higher institute admission are an intricate decision process and it is an important responsibility of the students to select the correct study track. The increase of the student's major dropout rate in higher education systems is one of the important problems in most institutions. One approach to solve such problem and succeed in academic life is to help the students in selecting a suitable major and assign them to the right track. The objective of our research is to build academic advisor model to students for their higher education which utilize classification data mining for recommending the suitable academic major. The method applied in the research is data mining classification techniques through decision tree method for advising students to select suitable major and help assign them to the right track. The proposed model classifies students and matches them to the proper study tracks according to their features. The three decision tree classification algorithms, namely J48, random tree and reduces error pruning (REP) tree was first applied to real data in a managerial higher institute in Giza Egypt and results are compared between them. Finally, the results showed that J48 algorithm gives 16 rules and we eliminate the rules that give low CGPA and we will use the 5 better rules that have the highest CGPA based on CGPA grade that equal (A) and J48 algorithm gives the highest accuracy 87.64% and classification error was 12.36% and was thus selected as the main classifier for building the proposed model based on the rules that we obtained from J48 algorithm than the two other classification algorithms and thus suggest using the generated J48 decision tree in our proposed student advising model to enhance students’ academic performance and decrease dropout.

Keywords : Data mining, Classification, Decision trees, Higher education, J48, Random tree, REP tree.
Cite this article : Mohamed Hegazy Mohamed and Hoda Mohamed Waguih, " A proposed academic advisor model based on data mining classification techniques " , International Journal of Advanced Computer Research (IJACR), Volume-8, Issue-36, May-2018 ,pp.129-136.DOI:10.19101/IJACR.2018.836003
References :
[1]Hand DJ. Principles of data mining. Drug Safety. 2007; 30(7):621-2.
[Crossref] [Google Scholar]
[2]Mahboob T, Irfan S, Karamat A. A machine learning approach for student assessment in E learning using quinlan s C4.5, naive bayes and random forest algorithms. In international conference on multi-topic. 2016 (pp. 1-8). IEEE.
[Crossref] [Google Scholar]
[3]Christian TM, Ayub M. Exploration of classification using NBTree for predicting students performance. In international conference on data and software engineering 2014 (pp. 1-6). IEEE.
[Crossref] [Google Scholar]
[4]Kumar SV, Padmapriya S. An efficient recommender system for predicting study track to students using data mining techniques. International Journal of Advanced Research in Computer and Communication Engineering. 2014; 3(9):7996-9.
[Google Scholar]
[5]Aulck L, Velagapudi N, Blumenstock J, West J. Predicting student dropout in higher education. ICML workshop on Data4Good: machine learning in social good applications, New York, USA 2016 (pp. 16-20).
[Google Scholar]
[6]Abu-Oda GS, El-Halees AM. Data mining in higher education: university student dropout case study. International Journal of Data Mining & Knowledge Management Process. 2015; 5(1):15-27.
[Crossref] [Google Scholar]
[7]Witten IH, Frank E, Hall MA, Pal CJ. Data mining: practical machine learning tools and techniques. Morgan Kaufmann; 2016.
[Google Scholar]
[8]Lakshmi Devasena C. Comparative analysis of random forest, REP tree and J48 classifiers for credit risk prediction. In international conference on communication, computing and information technology 2014 (pp. 30-6).
[Google Scholar]