Multimodal Learning Data Analysis and Algorithmic Teaching Effectiveness Evaluation Model Construction

Introduction:A combination of internal characteristics, such as motivation, personality, beliefs, and dispositions that combine with external circumstances to affect student results is referred to as teaching efficacy. The English-speaking environment becomes a major problem as students learn by interaction. To communicate effectively, they do not acquire the speaking and listening abilities necessary in English. Objective:This study examines the impact of computer programming courses and learning analytics on student’s computer programming skills. Methods:In this study, we proposed a novel Dung Beetle Optimized Flexible Random Forest (DBO-FRF) to evaluate the teaching effectiveness. In this study, 175 students’ data were collected for teaching effectiveness. Using student data, prediction model was constructed based on the attributes of the students, their past academic records, their interactions with online resources, and their advancement in laboratory work related to programming. The proposed method is compared to other traditional algorithms. Results and conclusion:The proposed method is implemented using Python software. The expected performance in the course and, in the instance that any submitted programs failed to meet the requirements, a programming recommendation from one of the class’s top students. The result shows the proposed method achieved better performance in terms of accuracy, precision, recall, and [Formula: see text]1-score. This decreased the performance gap between students who performed lower and those who performed higher, enabling students who adjusted their programs to learn more.