Recommender systems to support socio-collaborative learning in educational discussion forums

• Author / Creator

  Chen, Zhaorui

• With the popularity of online education, many educational technologies have been introduced to support students' learning. Among them, asynchronous discussion forums are widely used to support students’ socio-collaborative learning processes. However, the forum's complex thread structure and lengthy posts often lead to poor learning experiences: Students’ limited time is spent searching and filtering posts that match their interests. Accordingly, their time for engaging in more meaningful learning activities (i.e., discussion) is reduced. To address this issue, personalized learning support is needed. Forum post recommender systems are one of the possible solutions. However, none have been created for small-scale socio-collaborative learning forums that do not rely on a-priori domain knowledge and none integrate learning theories into the recommendation algorithm design. In this thesis, I introduce two similarly-structured multirelational graph-based recommender systems, CSCLRec and CoPPR. The recommender designs account for several learning theories and incorporate learner modeling, social network analysis, and natural language processing techniques. They customize forum post recommendations for learners with different social learning behaviors in order to accommodate individual learner needs in socio-collaborative online learning contexts. In the experiments with small course discussion forums, both CSCLRec and CoPPR delivered significantly better results than their competitors in terms of recommendation precision while achieving acceptable diversity and novelty performance. The results demonstrate that CSCLRec and CoPPR can predict students’ behavior and recommend relevant forum posts. These also imply the recommenders’ underlying ability to solve the information overload issue and increase student engagement in discussion-related learning activities.

• Subjects / Keywords

  • Discussion forum
  • Recommender system
  • Computer-supported collaborative learning
  • Online learning

• Graduation date

  Fall 2020

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-w2ed-ae12

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.