In today's higher education institutions, predictive analytics applications have become a pressing need. For all educational levels, predictive analytics used advanced analytics that included machine learning deployment to produce high-quality performance and insightful data. Most people are aware that a student's grade is one of the most important performance indicators that can aid educators in keeping track of their academic progress. Researchers have put forth numerous variations of machine learning approaches in the education area over the past ten years. To improve the accuracy of predicting student grades, addressing imbalanced datasets presents significant obstacles. Therefore, by enhancing the performance of predictive accuracy, this study gives a thorough analysis of machine learning algorithms to forecast the final student grades in the first semester courses. In this paper, two modules will be highlighted. Using a dataset of 1282 real student course grades, we assess the accuracy performance of six well-known machine learning approaches, including Decision Tree (J48), Support Vector Machine (SVM), Nave Bayes (NB), K-Nearest Neighbor (kNN), Logistic Regression (LR), and Random Forest (RF). In order to lessen the overfitting and misclassification results brought on by imbalanced multi-classification based on oversampling Synthetic Minority Oversampling Technique (SMOTE) using two features selection techniques, we secondly proposed a multiclass prediction model. The results collected demonstrate the suggested model's ability to integrate with RF and provide a significant improvement with the highest f-measure of 99.5%. In order to improve the prediction performance model for imbalanced multi-classification for student grade prediction, the suggested model shows comparable and encouraging results.
Components that we will provide are.
1.complete documentation support
2.complete working hardware/software implemented in students environment 3.classes will held accordingly.