<![CDATA[The graduation rate of students is an important measure of a school’s success, as it indicates the school’s ability to help students complete their education. Predicting student completion is crucial for schools to identify at-risk students and offer them early interventions to improve their academic performance. This can also assist policymakers in developing effective policies and programs to enhance graduation rates and reduce dropout rates. The dataset used in this study was obtained from the Kaggle website, and the best model proposed utilizes the Random Forest method with hyperparameter tuning. By adjusting the n_estimator parameter to 1000, our proposed method decreases the mean squared error (MSE) value from 0.5525155 to 0.5374983 and increases the R2 Score value from 0.9984039 to 0.9984873. The study also compares the performance of the proposed model with other datasets sourced from the University of California Irvine (UCI), demonstrating superior performance across all experiments. The results consistently show a decreasing trend in MSE value and an increasing trend in R2 value for all datasets]]>
Copyrights © 2024
