Evaluation of the Quality of Online Education Based on a Learning Interactive Network

Abstract

Contemporary education models are undergoing significant transformation, driven by the goals of ubiquity, intelligence, and personalization. This transformation is particularly evident in online education, exemplified by massive open online courses (MOOCs), which are increasingly becoming mainstream. The quality of online learning is now heavily influenced by interactivity, recognized as pivotal in supporting effective learning outcomes through enhanced help and feedback mechanisms. This study establishes an interaction network model for online learning, utilizing recurrent neural networks (RNNs) to embed learner and learning resource nodes into a Euclidean space. The primary aim is to evaluate the quality of interactions and determine whether they align with expected learning outcomes. The research methodology involves the development of an interaction network model and the application of RNNs for embedding nodes. Key metrics are proposed to assess interaction quality, integrating assessment feedback to enhance learning outcomes. Experimental validation on real-world datasets demonstrated the efficacy of the approach. The findings indicate that the proposed model significantly improves interaction quality in online learning environments. Recommendations include the adoption of similar interaction network models in educational platforms to optimize learning experiences. The implications of this study underscore the importance of robust interaction metrics in designing effective online learning environments. This study contributes to advancing the understanding and methodologies for modeling interactions in online education, emphasizing the critical role of interaction quality in achieving desirable learning outcomes.