Determining Student's Performance Effecting Factors to Predict Their Performances Using Various Classifiers

Abstract

Student performance in university courses is of great concern to higher education where several factors may affect the performance. This paper is an attempt to apply the data mining processes, particularly classification, to help in enhancing the quality of the higher education system by evaluating student data to study the main attributes that may affect the student performance in courses. For this purpose, we have used data obtained from Daffodil International University, Dhaka of Department CSE, batch 44. In this research, we will differentiate subjects taken by CSE students according to their performance and organize them. Then we will find the best features that count to student's performance using the information gain attribute of the Decision tree algorithm. Then we will find out the best machine-learning algorithm to predict the performance of students by comparing different classifier algorithms in both the holdout method and k-fold validation. Then we will discuss the impacts of our research on society.