in: Who Runs the World: Data, Sevinç Gülseçen,Emre Akadal,Sushil K. Sharma, Editor, Istanbul University Press, İstanbul, pp.145-163, 2020
Since the term “personalized learning” became popular, smart features have begun to be integrated into the e-learning environment. Data mining and machine learning algorithms are used to analyze big data stored in an e-learning system to make predictions to improve course quality or learners’ performance. From the learners’ perspective, it might now be considered possible for everybody to benefit from e-learning by considering their personal interests or their own specific development plan as long as the course contents are available in the system. In addition, in an e-learning environment, there is no limitation on the time and place where a course can be attended and a program completed. However, it is just not that simple. Today not the only, but by far the most important, the requirement is still the readiness of the learners to study in an e-learning system. The aim of this chapter is to predict the e-learning readiness of learners using data mining techniques. This chapter aims to provide feedback for institute managers and admin staff of e-learning systems which are intended to be used in an institution. Moreover, this section of the book contains one of the applications of big data analysis in education. Therefore, the main topic of this study is examined in terms of both classification and clustering techniques in order to provide a wider perspective to readers while using the sample application.
According to the results of this study, the highest accuracy value (0.831) is obtained with C4.5 Decision Tree Algorithm. While students, who agree and strongly agree with the statement “My studying/research area is appropriate for e-learning” are classified as ready to attend an e-learning course, students who disagree with the same statement are classified as not ready to attend an e-learning course. Students who strongly disagree with the statements “My studying/research area is appropriate for e-learning” and “E-learning is better than face to face learning”, are also classified as not ready to attend an e-learning course. Furthermore, the statement “My studying/ research area is appropriate for e-learning” is at the top of the obtained decision tree which indicates that it is an effective and directly related attribute which expresses student opinions about attending an e-learning course.