A Hybrid Data Fusion Model for Online Education Evaluation in Intelligent Classroom Environments
Abstract
The study uses a multi-source data fusion approach to assess online course effectiveness. The research combines student engagement metrics, learning performance, and course content quality using the 3P (Presage-Process-Product) model. The structured questionnaire was taken from undergraduate students from five regional colleges, and a total of 685 responses were collected, 583 of which were valid samples. The data were analyzed using SPSS 25.0 and AMOS 24.0 for reliability and validity testing. A backpropagation neural network (BPNN) was used to forecast learning performance related to course interaction, selection rates, and engagement metrics. Hypothesis testing was performed to check the links between course quality, engagement, and learning performance. Structural equation modeling was applied to find direct and indirect effects, leading to an unbiased approach for course effectiveness evaluation. The findings direct specific approaches to increase online course delivery, student engagement, and attainment of educational outcomes.