A learner’s role-based multi dimensional collaborative recommendation (LRMDCR) for group learning support
Xin Wan, Toshie Ninomiya, Toshio Okamoto · 2008
This article argues for the new solution of personal recommender systems that can provide learners with suitable learning objects to learn in group learning. In order to improve the ldquoeducational provisionrdquo to implement the e-learning recommender system, we propose a new recommendation approach which has been proven to be more suitable to realize personalized recommendation based on not only learning histories but also learning activities and learning processes which is defined as LRMDCR (a learnerpsilas role-based multidimensional collaborative recommendation) by us. In the approach, firstly we use the Markov chain model to divide the group learners into advanced learners and beginner learners by using the learnerspsila learning activities and learning processes. Secondly we use the multidimensional collaborative filtering to decide the recommendation learning objects to every learner of the group. We believe our approach is more effective and efficient to group learning.