WebCluster — Mediated Access to the Web

The WebCluster system enables structured, subject-specific portals to the World-Wide Web that provide a context for searching. A source collection of documents covering a given subject is automatically indexed, and the documents are clustered based on document-document similarity. The resulting hierarchical structure over the documents captures the topics covered in a subject area and acts as a subject-specific filter over the web. WebCluster has several applications where access can be mediated from any structured collection to any other searchable collection. A collection of documents covering a particular domain, e.g. nutrition, allows a structured content portal to be built that provides more effective access to both the original source collection, and to nutrition webpages. Existing information resources with a hierarchical structure, (e.g. large reports with many sections, structured websites or intranets, and structured electronic reference works) can be loaded by WebCluster, and access mediated through the structure.

Research Team

• David Harper (PI)
• Ayse Goker
• Bicheng Liu
• Mourad Mechkour
• Gheorghe Muresan

Related Projects

• iSee — supporting e-book comprehesion
• ProfileSkim — an intelligent document browser
• AmbieSense — ambient, personalised, and context-sensitive information systems for mobile users
• ProfileSkim — an intelligent document browser
• Facilitating Information Retrieval By User-Adaptive Learning
• Active Groups — knowledge sharing tools for email
• Personal Email Trainer (PET)

Publications

Harper, D. J., & Muresan, G. (2004). Mediated access to very large document collections. In Journal of American Society for Information Science and Technology. Liu, B., Harper, D., & Watt, S. (2004). Supporting federated information sharing communities. the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 572-573.

Muresan, G., & Harper, D. J. (2001). Document Clustering and Language Models for System-Mediated Information Access. In the Proceedings of Proceedings of the 5th European Conference on Research and Advanced Technology for Digital Libraries, pp. 438-449.

Muresan, G., Harper, D. J., Goker, A., & Lowit, P. (2000). ClusterBook, a tool for dual information access. In the Proceedings of 23rd Annual International SIGIR Conference on Research and Development in Information Retrieval, Athens, Greece, pp. 391.

Harper, D. J., Mechkour, M., & Muresan, G. (1999). Document clustering for mediated information access. In the Proceedings of 21st BCS-IRSG Annual Colloquium on IR Research, Glasgow, Scotland.

Mechkour, M., Harper, D. J., & Muresan, G. (1998) The WebCluster project: Using clustering for mediating access to the World Wide Web. In W. B. Croft, A. Moffat, C. J. van Rijsbergen, R. Wilkinson, and J. Zobel, editors, Proceedings of the 21st Annual International SIGIR Conference on Research and Development in Information Retrieval, Melbourne, Australia, 1998. ACM Press.