<![CDATA[This study aims to predict students’ academic scores using linear regression based on historical lecture data. The main problem addressed in this research is how academic data recorded during the learning process can be utilized to estimate students’ final scores in a more objective and measurable manner. This study employed a quantitative approach with a predictive research design. The analyzed data included assignment scores, quiz scores, midterm examination scores, attendance percentage, Learning Management System access, forum participation, late assignment submissions, and students’ final scores. A total of 180 valid student records were analyzed through data cleaning, descriptive analysis, correlation testing, multiple linear regression modeling, and model evaluation using R-square, Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, and Mean Absolute Percentage Error. The results show that the multiple linear regression model has strong predictive capability, with an R-square value of 0.784. The midterm examination score was identified as the most dominant predictor, followed by assignment scores, quiz scores, late assignment submissions, attendance, and LMS access. Forum participation showed a positive relationship but was not statistically significant. The evaluation results indicate that the model produced a relatively low prediction error, making it applicable as a basis for an academic early warning system. This study confirms that historical lecture data can be strategically used to support academic decision-making, monitor students at academic risk, and improve the quality of learning management in higher education]]>
