The Analysis of the Length of Studies in Higher Education based on Clustering and the Extraction of Association Rules

Abstract

The length of studies of the students who "linger" in Higher Education has not been justified in many countries, and the Higher Education Institutes try to solve the problem using various methods. The problem of students who "linger" in their Departments beyond the six or seven years is seen as complex one, in the Greek Higher Education. Two main alternative methods have been discussed: Giving the students who "linger" a low priority for registration in the laboratory classes, and limiting the number of times of attending laboratory based courses. Eventually, according to the new legislation the Greek Higher Education Institutes must cut off access to the students who "linger" too long. This study focuses on this hard problem. Clustering techniques and the mining of Association rules are used. The results of clustering and the generation of the association rules are based on students' questionnaires collected in the laboratory classes. Various interesting results and rules are extracted and discussed.

Developing Classification Model to Investigate the Problem of Computing Students Studies Length

“…The investigation about length of studies problem based on clustering and the extraction of association rules has been conducted in the Greek Higher Education by P. Belsis et al [3]. The students" data are collected through questionnaires in the lab classes.…”

“…In most higher education institutions, the length of study problem has not been investigated comprehensively despite the seriousness of this problem and its impact on the short andlong term. The students who beyond six or seven years in their academic program are called "linger" students [3], this definition exclude medicine domains.…”

Data mining techniques applied in educational environments: Literature Review

“…In 2014 we found papers in which data mining techniques were used, such as decision trees, clustering, association rules, neural networks and classification to address educational situations. The first one is "Data Mining: A prediction for Student's Performance Using Classification Method" (Badr, Din, & Elaraby, 2014) work developed by Ahmed A. and Elaraby I. where decision trees are used to predict the final students grade; on "Improving Quality of Educational Processes Providing New Knowledge using Data Mining Techniques" (Chalaris, Gritzalis, Maragoudakis, & Sgouropoulou, 2014) Chalaris M., Gritzalis S. and Maragoudakis M. use association rules to provide knowledge related to educational institutions processes; additionally, Belsis P., Chalaris I., Chalaris M., Skourlas C. and Tsolakidis A. published in "The Analysis of the Length of Studies in Higher Education based on Clustering and the Extraction of Association Rules" (Belsis, Chalaris, Chalaris, & Skourlas, 2014) how, from clustering and association rules extraction can analyze the study lenght in higher education; in the same year, Guruler H. and Istanbullu A. published "Modeling Student Performance in Higher Education Using Data Mining" (Mugla, 2014) where employ decision trees to identify factors that impact the success of students in higher education; also found a job Rabbany R.,Elatia S., Takaffoli M. and Zaiane O. entitled "Collaborative Learning of Students in Online Discussion Forums: A Social Network Analysis Perspective" (Rabbany, Elatia, Takaffoli, & Zaïane, 2014) where the clustering technique is used to analyze the social networks usage(ARS) and student interaction in forums; additionally, Yukselturk E., Ozekes S. and Türel Y. publish "Predicting dropout student an application of data mining methods in an online education program" (Yukselturk & Education, 2014) a writing where through the use of decision trees, neural networks and classification, examine the students dropout in online programs; in the same year, Chen X., Vorvoreanu M. and Madhavan K. published " Mining Social Media Data for Understanding Students' Learning Experiences" (X. Chen, Member, Vorvoreanu, & Madhavan, 2014a) in this paper with the use of classification technique analyze the students learning experience based on the information discussed in social networks; and finally, Hu Y., Lo C. and Shih S. in the paper entitled "Developing early warning systems to predict students' online learning performance" (Hu, Lo, & Shih, 2014) used classification and regression trees to create an early warning system of students performance in the LMS.…”

Association pattern of students thesis examination using fp-growth algorithms

“…The main goal of implementing data mining in the educational area is to use experience and new insight to improve the quality of education [10] and manage new courses properly [11]. Data mining also can use to prevent educational risk and educational opportunities i.e., student drop-out [12]- [16], duration of study [17], [18], learning behaviors [19]- [21], students outcome [22], [23] and student performance [10], [24]- [26].…”