Decision tree learning used for the classification of student archetypes in online courses

With ubiquitous Internet access nowadays, individuals have the ability to share more information than before, and it allows young people to collaborate and learn from a distance, so that educational systems are constantly being reshaped. Understanding eLearn-ing is important, and so is the typology of students who participate in this trend with increasing dedication. Yet, we consider that this accelerated pace of propagation of online education has left behind an important aspect needed for the act of teaching, namely studying and understanding student archetypes. By this we mean the common patterns which define the interaction type, dedication amount, and finalization perspective of courses. This paper introduces an original set of student profiles specific to online courses, and it does so by means of data mining and supervised learning. We use the responses from an online questionnaire to gather detailed opinion from 632 students from Romania regarding the advantages and disadvantages of MOOCs, as well as the reasons for not joining online courses. Based on the extracted statistics, we present six decision trees for classifying the finalization and participation rates of online courses based on the students individual traits. Furthermore, we discuss these profiles and explain the implications of this study. We believe our findings to bring consistent novelty both in understanding the needs of modern students, as well as in optimizing the way eLearning is further developed.