An Advisory Student Achievement Model Based on Data Mining Techniques

Abstract

Predicting student achievement is considered one of the most essential components of educational data mining. Academic institutions are concentrating on employing data mining techniques to improve student performance. Many prediction models have been presented to anticipate student accomplishment at an early stage in order to take preventative measures. This research looked at past models and presented a data-mining-based advising student achievement model. This study was carried out utilizing Artificial Neural Network (ANN), Decision Tree (DT), Naive Bayes classifiers, Random Forest, Support Vector Machine (SVM), and XGBoost to create a prediction model, with datasets containing 16 variables and 480 instances. Model produced satisfactory results, according to the findings of the experiments. With an accuracy of 84 percent without feature selection and 85.01 percent with feature selection using correlation, the XGBoost model was the most accurate of the four models.