An Ensemble Learning Approach for Extracting Concept Prerequisite Relations from Wikipedia

Online educational resources usually provided by different people, and the dependency relations between resources are always not clear, which bring challenges for self-directed learners, since they don’t know where to begin when they have a collection of resources. Concept prerequisite relations play an important role in educational resources sequencing and curriculum planning tasks. In this paper, we treat Wikipedia as an educational resource corpus and propose an ensemble learning approach for extracting concept prerequisite relations from Wikipedia. In experiments, we evaluate our approach on two existing datasets, the CMU and AL-CPL datasets, and validate that our approach can achieve better performance than baseline methods.