A Review on Student Attrition in Higher Education Using Big Data Analytics and Data Mining Techniques

Abstract

Student attrition among undergraduate students is among the most concerned issues in higher educational institutions in Malaysia and abroad. This problem arises when these students unable to complete their studies within the stipulated period when there are majoring in the Science, Technology, Engineering, and Mathematics (STEM) fields. Research findings highlight numerous factors contribute to the student attrition. These findings also suggest that the factors differ from one case to another case. Effects of student attrition not only for the student itself but also to the institutions and community. It is challenging to classify the factors based on general assumptions. Moreover, increasing students' information makes the problem more complicated. This student information can provide a useful database for analytical analysis. Methods such as big data analytics and data mining techniques can be deployed to gain insights and pattern that related to student attrition problem. The objective of this paper (i) review the student attrition in higher education (HE) and the contributing factors; and (ii) review the existing computational model to analyze and predict student attrition in HE.

Deep Learning Model to Predict Students Retention Using BLSTM and CRF

“…Terms relevant to a lack of student retention include student withdrawal, student attrition, and student dropout. These terms refer to the circumstance where the student has been unable to complete his/her studies and has subsequently left the course prior to its completion [12]. The complex interrelationship of factors influencing student retention in higher education is also increasingly influenced by advancements in both information and communication technologies and the capabilities of universities to collect, manage, and analyze student enrolment data.…”

A Comparative Analysis of Clustering Algorithm

“…k-mean is intuitive and easy to implement, however, it has some drawbacks. Centroid Selection process [1] has to be very precise and knowledge about actual Clusters is mandatory for k-Means algorithm which is difficult to predict in dataset pertaining to live cases. The kmedoid algorithm works on subset of points, called medoids, such that average dissimilarity between them and proximity.…”

A Phenomenological Study on International Student Withdrawals from Malaysian Higher Education Institutions

“…The studies done so far were focusing on local students or specific group of students in primary, secondary and high school levels, and post-secondary levels which include colleges, vocational and university students. Furthermore, several studies have attempted to build models to evaluate and predict student dropout or at students at risk of dropping out using meta data (Rovira, Puertas, & Igual, 2017;Sangodiah, Beleya, Muniandy, Heng, & SPR, 2015;Tarmizi, Mutalib, Hamid, & Rahman, 2019;Von Hippel & Hofflinger, 2020). Such studies have mixed local (domestic) and international students with a data mining approach.…”