<![CDATA[The increasing number of students and the growing adoption of digital services have made campus service demand more dynamic, creating a need for accurate prediction methods to support service planning and data-driven decision-making. This study aims to analyze and compare the performance of Neural Network and Linear Regression algorithms in predicting campus service demand based on student data at AMIK Medicom. The dataset consists of student-related variables, including the number of active students and the percentage of digital service utilization as predictor variables, while total campus service demand is used as the target variable. The data were collected from September 2025 to February 2026. The research procedure involved data collection, data preprocessing, the development of Linear Regression and Neural Network models, and model evaluation using the coefficient of determination (R²), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and prediction accuracy. The results indicate that both methods are capable of predicting campus service demand effectively. The Linear Regression model achieved an R² value of 0.987, an MAE of 245.60, an RMSE of 301.20, and an accuracy rate of 91.4%. Meanwhile, the Neural Network model achieved an R² value of 0.996, an MAE of 118.30, an RMSE of 156.50, and an accuracy rate of 96.8%. These findings demonstrate that the Neural Network model outperforms Linear Regression, as evidenced by its higher R² value and lower error rates. Therefore, Neural Network is recommended as a more effective prediction model for supporting data-driven planning and management of campus services at AMIK Medicom.]]>
