Prediction of Student Performance in Higher Academy Through Different Data Mining Techniques

Providing a quality education to its students and reducing dropout rates are two of the primary goals of higher education. This can be done by forecasting student performance using educational data mining (EDM). This study aims to measure the performance of students in the different categories. The categories are excellent, good, moderate, poor, and probation, and the scores are compared between excellent, poor, and probation students using some effective features. In this work, the data was gathered from 675 records of undergraduate students from the computer science and engineering departments of a renowned private university in the city of Dhaka, Bangladesh. Finally, recommendations are given to learners on how to improve performance to achieve academic success, and the following six data mining methods are used for accuracy: K-Nearest Neighbor, Decision Tree, Gaussian Naive Bayes, Support Vector Machine, Logistic Regression, and Random Forest.