Clustering of Participation Degrees of Distance Learning Students to Course Activity by Using Fuzzy C-Means Algorithm

Abstract

This study presents a clustering method based on Fuzzy C-Means algorithm for determining the participation degree of distance learning students to course activity. It will be used to predict academic performance of the students at the end of the semester. The course activity data of 6659 students who took the course of Information and Communication Technologies given through distance learning in Mugla Sitki Kocman University within the last three years were selected as the sample in the study. The data set consists of how often students visited the course page, how often they watched the course materials, how long students watched the course videos, and how long the participants attended online classes after entering the Moodle learning management system. In the KNIME data mining platform, we analyzed the data set which is cleaned from noisy data. As a result of the analysis, raw data is automatically transformed into fuzzy clusters. The fuzzy clusters obtained as a result of the study provide instructors with information about determining the performance score of the students at the end of the year.