Detecting Special Lecturers Using Information theory-based Outlier Detection Method
Thuc-Doan Do, Minh Tuan Tran, Xuan-May Thi Le, Thuy‐Van T. Duong · 2017
The faculty evaluation forms can be considered as valuable data source to exploit knowledge that helps to improve the quality of teaching and learning in universities. In this paper, we analyze previous studies on exploiting faculty evaluation forms and outlier detection task in data mining. On that basis, we propose and solve the problem of detecting special lecturers assessed considerably differently from the remainder using the efficient information theory-based algorithm in [14]. The data are collected from the online faculty evaluation system of our university, with more than 140,000 evaluation forms. Experimental results show that our solution is both effective and efficient. It is much faster than the fast greedy information theory-based algorithm in [13] while their accuracies are competitive with each other.