Multilayer knowledge graph construction and learning behavior routing guidance based on implicit relationships of MOOCs

Abstract

As a mature model of open or distance education, MOOCs provide learners and teachers with flexible knowledge propagation and independent learning methods, facilitate the online resource sharing of teachers, learners and knowledge, and realize the digital management and application of entire learning processes, but also brought the key challenge, how to achieve the adaptive learning behaviors based on knowledge graphs? This study constructs one multilayer knowledge graph prediction model of MOOCs, and implements the implicit relationships to drive suitable learning behavior routes. Firstly, the knowledge graphs in MOOCs are analyzed and described; Secondly, we combine with data structures and relationships, define appropriate node semantics and relationship characteristics, as well as learning behavior scenarios and learning needs, PMIR-BG model based on improved bipartite graph is designed to predict the implicit relationships. Experiments show that the model is efficient and reliable for tracking implicit relationships and building multilayer knowledge graphs; Thirdly, the analysis conclusions of corresponding problems are demonstrated. Through visualizing the knowledge graphs of multilayer relationships, the learning behavior routing guidance is realized. The whole research has strong theoretical value and practical significance, and provides key technologies and decisions to construct and optimize the personalized and adaptive learning processes.