Automatic extraction of prerequisites and learning outcome from learning material

摘要

To provide an adaptive guidance to the student through learning object repositories, a system needs to have knowledge about the learner as well as material. This paper discusses identification of metadata that describes the pedagogical aspects of a document. Learning goals and prerequisites, if stored as part of metadata, can be utilised for intelligent recommendations. Manual annotation is a time consuming and expensive process. Correct instantiation of learning object metadata requires combined educational and technical skills. This paper proposes natural language processing-based automatic concept extraction and outlines rule-based approach for separation of prerequisite concepts and learning outcomes covered in learning document. The importance of automatic extraction of prerequisite terms is strengthened after evaluation of this work, since results show that concepts missed by authors are suggested by the system. To increase precision of retrieval, subject domain ontology is used.