Relevance feedback versus web search document clustering

The performance of an IR system is deteriorated by factors including short and vague queries put up by the users, ever increasing volume of documents on the web, users not knowing their exact information need etc. Relevance feedback (RF) and web search document clustering are techniques to improve the performance of an Information Retrieval (IR) system. Relevance feedback provides a method to get more relevant search result from an IR system using documents that are marked relevant by the user as a feedback to reformulate query. This refined query is then used to retrieve the documents. In document clustering approach, the search result is divided into thematic groups where documents of one group are similar to each other and dissimilar to the documents of other groups. This paper presents a report on the effectiveness of relevance feedback technique as compared to document clustering in context of web information retrieval and why document clustering is the most preferred approach.