Attributes of Low Performing Students In E-Learning System Using Clustering Technique

Abstract

Data mining in education is considered to be one of the relevant and fast growing areas in data mining, with free access to datasets available online, researchers have continued to analyze and produce knowledge which has improved the educational sector. With many research geared towards predicting student results, this paper offers a different approach of gaining knowledge of student data by presenting the attributes of low-performing students. The idea is to group students with low grades and discover the core attributes of these category of students, thereby providing stakeholders with these attributes which should be looked out for in current and prospective students. The dataset used in this research was collected from an e-Learning system called Kalboard 360. The k-means clustering technique embedded in the WEKA tool was used to group these category of students into two clusters. The knowledge gained from the mining process shows that lower-level absentee students with parents that do not actively participate in their learning process are most likely to perform poorly in their studies.