A dynamic Hebbian learning algorithm for constructing e-learner communities

Authors:
Qinghua Chen;Jing Jin;Huaxi Chen
Affiliations:
School of Electroincs and Information Technology, City College, Wenzhou University, Wenzhou, China;Dept. Software Research & Development, Microsoft Research Asia, Beijing, China;Dept. Computer Science and Technology, Shanghai Jiao Tong University, Shanghai, China
Venue:
ICNC'09 Proceedings of the 5th international conference on Natural computation
Year:
2009

Citing 2
Cited 0

Search and replication in unstructured peer-to-peer networks

ICS '02 Proceedings of the 16th international conference on Supercomputing
An Intelligent Knowledge Sharing Strategy Featuring Item-Based Collaborative Filtering and Case Based Reasoning

ISDA '05 Proceedings of the 5th International Conference on Intelligent Systems Design and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to solve the problem of "information overflow" in e-learning, an algorithm based on Hebbian learning law is proposed for constructing self-organized communities which can automatically group e-learners according to their learning interests. Unlike filtering methods, this algorithm takes into consideration of the distributed open environment of e-learning. This paper designed a peer-to-peer architecture and applied Hebbian learning law in constructing e-learner communities, avoiding the difficulty in calculating user similarity. Compared with traditional Hebbian learning based algorithm, this algorithm uses dynamic thresholds to address the problem of unilateral trust weight adjustment in extreme cases, and it also improves the trust weight adaptation and neighbor adjustment policies. Experimental results show that this algorithm achieves faster community construction speed and better scalability and stability.