Research pick: Testing times - "Automatic generation of questions from DBpedia"

Writing quiz-type tests in education can be time-consuming and given the nature of education and home learning during the current coronavirus pandemic, teachers and carers alike need more efficient and straightforward ways to produce quizzes to evaluate learning in their student charges.

New work published in the International Journal of Control Engineering Education and Life-Long Learning suggests that quiz questions could be generated automatically from DBpedia. DBpedia is a project that extracts structured content from Wikipedia and makes it available on the World Wide Web. Users can semantically query relationships and properties of Wikipedia resources, including links to other related datasets.

Oscar Rodríguez Rocha, Catherine Faron Zucker, Alain Giboin, and Aurelie Lagarrigue of the Universite Cote d’Azur, in France, explain that test questions must be generated in compliance with the knowledge and skills necessary to master a specific subject and be appropriate for the school year being tested and in accord with official educational standards. They have now demonstrated how pertinent information can be pulled from a structured database that fits these criteria. Question-generating methods can then be applied automatically to the knowledge scraped in this way to produce the questions and answers for a school quiz.

The educator can curate the information and questions manually to ensure they are wholly appropriate to the students. The automated part of the system can reduce the educator burden considerably freeing up time for less mundane work with the students while at the same time providing an invaluable and validated assessment tool.

Rodríguez Rocha, O., Zucker, C.F., Giboin, A. and Lagarrigue, A. (2020) ‘Automatic generation of questions from DBpedia’, Int. J. Continuing Engineering Education and Life-Long Learning, Vol. 30, No. 3, pp.276–294.