Student Grade Prediction and Classification based on Term Frequency-Inverse Document Frequency with Random Forest

In recent years, the academic grade prediction has obtained interest in education to improve the student employment opportunities. Where developing automatic grades to the students in online becoming a major issue. The existing Random Forest (RF) method is utilized for prediction. In this paper, the Term Frequency-Inverse Domain Frequency with Random Forest (TF-IDF-RF) is proposed for student grade prediction and classification with higher education student performance evaluation dataset. Then, the data is pre-processed with various techniques such as data cleaning and data categorization and data reduction to enhance the input data for the effective performance of the model. After that, this preprocessed data is further processed to feature extraction with Term Frequency-Inverse Document Frequency (TF-IDF) to convert the entire data into vectors for better classification. Furthermore, processed to classification to classify the student performance based on their grades additionally prediction is also done with RF model. From the results, the proposed TF-IDF-RF model achieved better results when compared to existing KNN and results in terms of accuracy, precision, recall and f1-score of ${9 9. 5 \%}, {9 5. 2 \%, 9 6. 4 \%}$ and ${9 5. 8 \%}$ respectively.