Methodological Implementation for Predicting Student Performance Using Data Mining Classifiers and Machine Learning

Each engineering university's major goal is to provide students with a world-class education. However, following COVID-19, when schools reopened, there was a large announcement that pupil overall performance was deteriorating daily because everything had changed from online to offline. Predicting negative grades among college students at the beginning level is critical so that their professors and management can assist and encourage them to change the negative grade into a top grade. My research area comes and provides strategies to reduce academic troubles among college students. The key goal of this research study is to become acquainted with the many aspects that aid in forecasting learner success. The second purpose is to examine and analyze the various facts. Data Mining classifiers and Artificial Neural Networks on a dataset including 689 student records of IP University for prediction of student performance in Python programming language. This work then entails calculating the Accuracy, Fl-score, Recall value, Precision value, ROC curve, and the Confusion matrix of several algorithms and determining the best set of rules based on the following metric scores. A Random Forest was executed with 98% accuracy.