IROPINE Seminar: A Multi-level Recommender System for E-learning


Time & date:
11:00 am – 12:00 noon, 30 December 2016 (Friday)
Venue:
C0614, 6/F, Block C, Main Campus, OUHK
Language:
English / Cantonese
Speaker:
Professor Philips Fu Lee Wang
Vice-President (Research and Advancement)
Caritas Institute of Higher Education

Professor Philips Fu Lee Wang is the Vice-President (Research and Advancement) at Caritas Institute of Higher Education. He received PhD from The Chinese University of Hong Kong. His research interests include electronic business, information retrieval, information systems and e-learning. He has been Principal Investigator of 12 grants, Co-Principal Investigator of 3 grants, and Co-Investigator of 9 grants. The total funding of projects is over HK$220M. He has over 200 publications, including IEEE Intelligent Systems, Decision Support Systems, Journal of the American Society for Information Science and Technology, Neurocomputing, Information & Management, Information Processing & Management, IEEE Multimedia, International Journal of Machine Learning and Cybernetics, Journal of Intelligent and Fuzzy Systems, Information Processing Letters, and more. He is Past Chair of ACM Hong Kong Chapter and IEEE Hong Kong Section Computer Chapter. He is also Fellow Members of Hong Kong Institute of Chartered Secretaries and Institute of Chartered Secretaries and Administrates.

Abstract:
With the rapid development of MOOCs and other e-learning applications and platforms, the volume of e-learning resources has grown significantly in recent years. At the same time, a variety of modalities of learning resources are provided. The fast growth of e-learning resources results in the problem of information overload as learners are overwhelmed by the rich learning resources. In other words, it is increasingly difficult for learners to find required learning materials efficiently and effectively. We propose a powerful framework to organize e-learning resources and capture learning preferences by integrating various hidden relationships among learners, knowledge units and resources. Subsequently, the learning resources will be recommended at different levels. Firstly, learning resources are recommended to individual learner based on the contextualized learner profile so that personalized learning could be facilitated. Secondly, group member discovery could be achieved by assigning learners into groups based on their distance from contextualized group profiles. Thirdly, learning resources can be recommended to a class based on the aggregated class profile. Throughout the case studies, we have verified that the proposed framework is very flexible and powerful to support various kinds of e-learning applications in different scales.
 

* Compulsory field
PERSONAL INFORMATION


OTHER INFORMATION
1. * OUHK Unit (e.g. A&SS, LiPACE, SAO, ETPU)



2. * Position (post title)