Construction of a knowledge graph for exercise selection

A. Bijlsma, Cornelis Huizing, A.J.F. Kok, Ruurd Kuiper, H.J.M. Passier, Erik Scheffers, S. Schivo, T.E.J. Vos

Research output: Book/ReportTechnical reportAcademic

67 Downloads (Pure)

Abstract

This paper describes the construction of an open-source repository to store and select exercises, whose architecture and design are usable for many knowledge areas. The repository has been built and is operational for a collection of Java programming exercises, independent of any specific teaching approach. Each exercise is equipped with a set of knowledge items to clearly show which programming knowledge is needed. Consequently, a tag-based search enables students to find those exercises that match their knowledge level. In order to make the process of tagging and searching efficient, the exercise tags have been structured into a foreknowledge graph. In this graph, each node is a tag while edges describe dependency relations between them. An interactive online tool is provided that supports the submission of and search for exercises.
Original languageEnglish
PublisherOpen Universiteit
Number of pages24
Publication statusPublished - 14 Oct 2021

Publication series

SeriesOUNL-CS (Technical Reports)
Number2
Volume2021

Keywords

  • Knowledge graph
  • exercise
  • repository

Fingerprint

Dive into the research topics of 'Construction of a knowledge graph for exercise selection'. Together they form a unique fingerprint.

Cite this