Predicting success in Computer Engineering at University of Rwanda using machine learning

  • Jennifer Batamuliza
Keywords: Computer, Engineering, Programming, Language, Education, Machine, learning, Conceptual, prediction


This research presented an educational software or instrument for predicting student success at University of Rwanda in Computer Engineering. In most universities, success is a vital concern certainly in programming environment. Therefore, this research indicated variables that can be used
in prediction of success in computing area. The research focused on characteristics like student’s mathematical background, programming aptitude, problem solving skills, gender, and prior experience, previous computer programming experience, and e-learning usage which can help in the
analysis of student’s success within or outside the university. Predictive modeling is very effectively implemented in student’s success and the most frequent methodologies are the Decision Tree algorithm and the Regression Algorithm. The user will provide student’s details, then inputted
values will be collected and the system will generate the result based on the input using different algorithms. In reference to this system’s objectives, a fully software or instrument for predicting system was developed using the aforementioned tools. As though a few numbers of previous works available for students result prediction but there are fewer variables, which are used to predict success and the most important is for example the attendance rate. This project work is used to know the academic status of the students. Also it is easily used to know how much the student was absent or present, how far the student is involved in activities of the colleges or schools. Algorithms such as decision tree classifier, linear regressions and many more are used.

Key Words: Computer, Engineering, Programming, Language, Education, Machine, learning, Conceptual, prediction .


Journal Identifiers

eISSN: 1597-4316