CLASSIFICATION OF STUDENTS’ ACHIEVEMENT VIA MACHINE LEARNING BY USING SYSTEM LOGS IN LEARNING MANAGEMENT SYSTEM

During emergency remote teaching (ERT) process, factors affecting the achievement of students have changed. The purposes this study are to determine variables that affect classification according their course achievements in ERT during pandemic process and examine performance machine learning techniques. For these purposes, logs from management system were used. In study, analyzes carried out with various techniques performances compared. As a result it was observed Fisher’s Linear Discriminant Analysis best technique F measure criteria. another result, most effective variable, classifying students, is average number days logged into per month week. It has been total activity duration (min), weeks page views semester less influential factors. Accordingly, could be suggested check monthly weekly follow-up lectures instead follow-ups semester. addition, students’ interaction patterns can monitored tracking systems.