Authors: H. Hannah Inbarani; K. Thangavel
Addresses: Department of Computer Science, Periyar University, Salem-636 011, Tamil nadu, India ' Department of Computer Science, Periyar University, Salem-636 011, Tamil nadu, India
Abstract: Web personalisation systems include a new generation of recommender systems that integrate multiple online channels, are more scalable, are more adaptive and can better handle user interactivity. Efficient and intelligent techniques from artificial intelligence, machine learning, web mining and statistics are needed to mine this data for actionable knowledge, and to effectively use the discovered knowledge to enhance the users' experience. In this paper, we propose a rough biclustering approach for creating user model, based on which user profiles are constructed and the user profiles are matched with the active user session for web page recommendation. To determine the effectiveness of the proposed approach, it is compared with conventional biclustering, spectral co-clustering and CDK-means biclustering using the evaluation metric used for page recommendation. The experimental results show that the proposed rough biclustering algorithm outperforms the other approaches for web page recommender systems.
Keywords: co-clustering; rough biclustering; web personalisation; webpage recommendation; user modelling; recommender systems; user profiles; artificial intelligence; machine learning; web mining; statistics.
International Journal of Granular Computing, Rough Sets and Intelligent Systems, 2013 Vol.3 No.1, pp.59 - 84
Available online: 22 May 2013 *Full-text access for editors Access for subscribers Free access Comment on this article