<![CDATA[This study aims to predict the semester in which students complete their final thesis using transcript data and three regression algorithms: Linear Regression, Kernel Ridge Regression, and Decision Tree Regression. The research evaluates the performance of each model using Mean Squared Error (MSE) and Mean Absolute Error (MAE) as evaluation metrics. The experimental results show that Kernel Ridge Regression outperforms the other two models with an MSE of 2.271 and an MAE of 1.251. In comparison, Linear Regression achieved an MSE of 5.137 and an MAE of 1.859, while Decision Tree Regression produced an MSE of 4.1 and an MAE of 1.2. These findings indicate that Kernel Ridge Regression is the most effective method for predicting the completion semester based on academic transcripts, providing more accurate and reliable results. The study contributes to the academic field by demonstrating the potential of machine learning models in predicting students' academic progress and supporting better decision-making for academic management.]]>
