Comparison of Different Classification Techniques for Educational Data

Literature Review

Educational Data Mining (EDM) is a relatively new stream in the data mining research and there are only a few studies by researchers in this field. Extensive literature reviews of the EDM research field have been done by Romero & Ventura (2007), covering the research efforts in the area, between 1995 and 2005, and by Baker & Yacef (2009), for the period after 2005. Romero & Ventura surveyed the application of data mining to traditional educational systems, particularly web-based courses, well-known learning content management systems, and adaptive and intelligent web-based educational systems. They have applied data mining techniques viz. statistics and visualization; clustering, classification and outlier detection; association rule mining; pattern mining and text mining.

The potential applications of data mining in higher education have been explained by Luan in his work in 2002. The author has also discussed how data mining saves resources while maximizing efficiency in academics. Ma, Liu, Wong, Yu, & Lee (2000) focussed on understanding student types and then opting for targeted marketing by applying different data mining models. Tair & El-Halees (2012) have used educational data mining to improve graduate students’ performance, and to overcome the problem of low grades of graduate students. The authors have tried to extract useful knowledge from graduate students data collected from the college of Science and Technology for a period of fifteen years [1993-2007].