Current Proceedings on Technology

Yazarlar: Pragya Dwivedi, Kamal K. Bharadwaj

Konular:-

Anahtar Kelimeler:Elearning,Recommender System,Collaborative Filtering,Learning Styles,Nowledge Levels

Özet: E-Learning recommender system is an important information exploration paradigm that retrieves relevant learning resources to learners based on their preferences expressed implicitly or explicitly while studying using e-learning systems. The work presented in this paper is an attempt towards addressing the problem of recommending learning resources to facilitate a group of learners in e-learning environment. We analyze the desiderata of group recommendations for learners and propose an effective group recommendation procedure that improves not only the group recommendation effectiveness but also the satisfaction of individual group members. First of all, we generate individual resource satisfaction to each member of a group by employing collaborative filtering framework utilizing both the learning styles and the knowledge levels. Thereafter, we formalize the problem of group recommendation and construct the group consensus function that aims at maximizing resource satisfaction as well as agreement between group members. Experimental results are presented to demonstrate the effectiveness of proposed group learning recommender system.

---