Mining Evolving Web Sessions and Clustering Dynamic Web Documents for Similarity-Aware Web Content Management

摘要

Similarity discovery has become one of the most important research streams in web usage mining community in the recent years. The knowledge obtained from the exercise can be used for many applications such as predicting user's preference, optimizing web cache organization and improving the quality of web document pre-fetching. This paper presents an approach of mining evolving web sessions to cluster web users and establish similarities among web documents, which are then applied to a Similarity-aware Web content Management system, facilitating offline building of the similarity-ware web caches and online updating of sub-caches and cache content similarity profiles. An agent-based web document pre-fetching mechanism is also developed to support the similarity-aware caching to further reduce the bandwidth consumption and network traffic latency, therefore to improve the web access performance.