Research on multimodal based learning evaluation method in smart classroom

In traditional learning contexts, teachers primarily assess students' behavior, emotional changes, and assignment completion to ensure teaching quality. Currently, there are challenges in evaluating students, such as assessments being insufficiently comprehensive and timely, a singular evaluation perspective that hinders the holistic consideration of factors affecting learning assessments, and a weak correlation among evaluation criteria, resulting in suboptimal evaluation outcomes. In recent years, with the rapid development and widespread application of artificial intelligence and information technology, the era of smart classrooms has arrived. New technologies like image processing and artificial intelligence offer opportunities for personalized support services and enhancing teaching quality. Therefore, to provide a more comprehensive and objective reflection of teaching quality, this paper proposes a multi-modal information fusion learning assessment model. This model is achieved by determining the weight values of three dimensions, cognitive attention, emotional attitude, and course acceptance along with their corresponding attributes. Subsequently, through a fusion strategy, it calculates the learning assessment score by integrating information from these three dimensions. A series of experimental data confirms the effectiveness of this approach.