<![CDATA[The aim of this study was to compare the effectiveness of Project-Based Learning (PBL) and Lecture-Based Learning (LBL) in enhancing students’ ability to implement machine learning models. While both teaching methodologies are widely used, the impact of PBL on practical machine learning skills has not been sufficiently explored. This study investigates whether a hands-on, project-based approach leads to better performance in real-world machine learning applications compared to a traditional lecture-based approach. This experimental study involved 60 undergraduate students from a machine learning course, randomly divided into two groups: PBL and LBL. The PBL group worked on real-world machine learning projects, while the LBL group followed traditional lectures and individual assignments. Data were collected using pre- and post-test questionnaires, project performance rubrics, and observational notes. Statistical analyses were conducted to compare the two groups’ performance on machine learning tasks. The results revealed that the PBL group outperformed the LBL group in model accuracy, code quality, problem-solving, and debugging skills. The PBL group also demonstrated greater motivation and engagement, with statistically significant differences in performance (p = 0.001 for model optimization and p = 0.027 for problem-solving). The LBL group showed improvements, but the gains were less substantial. The findings suggest that PBL is more effective for developing practical skills in machine learning. However, the study's limitations include a small sample size and short duration, which may limit the generalizability of the results. This study provides novel insights into the benefits of PBL in machine learning education, offering valuable implications for curriculum design. Future research could explore long-term outcomes and the potential of hybrid teaching methods that combine PBL and LBL]]>
Copyrights © 2025
