Abstract—This paper discusses how to generate question templates by interpreting existing assessment questions from textbooks and validating concepts from existing ontologies. Previous work shows that most of the concepts that exist in the ontologies can be used as keywords to generate useful assessment questions. In this paper, we consider questions generated from a variety of ontologies, and question taxonomies, which include definitions, concept completions, and comparisons. Evaluation of novel techniques for question generation provides insights on how to extract a pattern from assessment questions in a textbook in order to create question templates that can be used with concepts presented in ontologies to generate useful questions. This paper discusses the method used and the experimental results.
Index Terms—Question template, question generation, domain ontology.
N. H. Ibrahim Teo and M. S. Joy are with the Department of Computer Science at the University of Warwick, Coventry, CV47AL, UK, University Teknologi MARA, Jasin Melaka, Malaysia (e-mail: shimateo@uitm.edu.my, N. H. Ibrahim-teo@warwick.ac.uk, m.s.joy@warwick.ac.uk).
[PDF]
Cite: Noor H. Ibrahim Teo and Mike S. Joy, "Categorized Question Template Generation for Ontology-Based Assessment Questions," International Journal of Knowledge Engineering vol. 4, no. 2, pp. 72-75, 2018.