Contexts in a paper recommendation system with collaborative filtering

Authors

  • Pinata Winoto Konkuk University, Chungju City
  • Tiffany Y. Tang Konkuk University, Chungju City
  • Gordon I. McCalla University of Saskatchewan

DOI:

https://doi.org/10.19173/irrodl.v13i5.1243

Keywords:

e-learning, pedagogy

Abstract

Making personalized paper recommendations to users in an educational domain is not a trivial task of simply matching users’ interests with a paper topic. Therefore, we proposed a context-aware multidimensional paper recommendation system that considers additional user and paper features. Earlier experiments on experienced graduate students demonstrated the significance of this approach using modified collaborative filtering techniques. However, two key issues remain: (1) How would the modified filtering perform when target users are inexperienced undergraduate students who have a different pedagogical background and contextual information-seeking goals, such as task- and course-related goals, from those of graduate students?; (2) Should we combine graduates and undergraduates in the same pool, or should we separate them? We conducted two studies aimed at addressing these issues and they showed that (1) the system can be effectively used for inexperienced learners; (2) recommendations are less effective for different learning groups (with different pedagogical features and learning goals) than they are for the same learning groups. Based on the results obtained from these studies, we suggest several context-aware filtering techniques for different learning scenarios.

Author Biographies

Pinata Winoto, Konkuk University, Chungju City

Assistant Professor Department of Computer Engineering

Tiffany Y. Tang, Konkuk University, Chungju City

Assistant Professor Department of Computer Engineering

Published

2012-11-01

How to Cite

Winoto, P., Tang, T. Y., & McCalla, G. I. (2012). Contexts in a paper recommendation system with collaborative filtering. The International Review of Research in Open and Distributed Learning, 13(5), 56–75. https://doi.org/10.19173/irrodl.v13i5.1243