Student attendance in lectures plays a crucial role in academic achievement and the quality of learning. The Decision Tree algorithm is used to analyze student attendance patterns with a dataset containing 6,607 entries from Kaggle, comprising 20 related attributes. Using RapidMiner, the analysis process includes data splitting, model building, and performance evaluation. The model achieved 49.96% accuracy, with the best performance in the "Medium" class (50.40% precision, 98.12% recall) but showed weaknesses in the "High" and "Low" classes. These results highlight the importance of data-driven approaches to designing effective strategies, such as rescheduling or improving teaching methods, to enhance student participation.