Exploration of Student Performance through the Application of Data Mining

Abstract

Using data mining in the education business to estimate the performance of students who are enrolled in universities is the topic of this paper, which presents the findings of a study. Both of these data mining algorithms are utilized. A descriptive assignment that is based on the K-means algorithm is used to pick multiple groups of pupils as the first step. Second, a classification job assists with two classification techniques, namely Decision Tree and Naïve Bayes, which are responsible for predicting students who drop out of college due to poor performance in the first four semesters of their college attendance. Cross-validation techniques were utilized to evaluate the models, which were trained and tested using the academic data collected from students throughout the admissions process. After adding data from past academic enrollment, the findings of the experiment demonstrate that the prediction of students dropping out of school improves, and that student performance is monitored.