Tracking Student Performance in Introductory Programming by Means of Machine Learning

large amount of digital data is being generated across a wide variety of fields and Data Mining (DM) techniques are used transform it into useful information so as to identify hidden patterns. One of the key areas of the application of Education Data Mining (EDM) is the development of student performance prediction models that would predict the student's performance in educational institutions. We build a model which can notify students (in introductory programming course) about their probable outcomes at an early stage of the semester (when evaluated for 15% grades). We applied 11 Machine Learning algorithms (from 5 categories) over a data source using WEKA and concluded that Decision Tree (J48) is giving higher accuracy in terms of correctly identified instances, F-Measure rate and true positive detections. This study will help to the students to identify their probable final grades and modify their academic behavior accordingly to achieve higher grades.