SVD: a novel content-based representation technique for Web documents

Most users typically express their information need via short queries to search engines and they often have to physically sift through the search results based on relevance ranking, making the process of relevance judgement time-consuming. The structure of the Web is increasingly being used to improve organisation, search and analysis of information on the Web. In this paper, we describe how the Web structure together with our novel summarisation techniques can be applied to better represent knowledge in actual Web documents via the proposed semantic virtual documents (SVD). We will also outline our experimental design to evaluate the effectiveness of the proposed SVD with a prototype system called iSEARCH (intelligent search and review of cluster hierarchy) for Web content retrieval and mining. The experimental results confirm that the novel technique show promising qualities in content-based representation for Web documents to enhance Web content retrieval and mining.