Web search based on web communities feedback data

Abstract

The capability to easily find relevant information becomes increasingly difficult as the available content increases. Web Search Engines aim to assist users in finding pertinent information. To measure the relevance of a Web page (its rank), different strategies are used. However, page ranking is mainly conducted by relying on automatic assessment criteria. Hence, a gap is created between the effective relevance of a content and the computed one. To reduce this gap, we introduce a framework for feedback-based web search engine development. To illustrate the effectiveness and the use of the proposed framework, we developed a web search engine prototype called SocialSeeker. Finally, we evaluated our approach from the end-user perspective and the results shown that feedback data can improve search engine results. © 2009 Springer Berlin Heidelberg.