A Particle Swarm Optimization Algorithm for Web Information Retrieval: A Novel Approach

Abstract

The Web has become the largest source of information worldwide and the information, in its various forms, is growing exponentially. So obtaining relevant and up-to-date information has become hard and tedious. This situation led to the emergence of search engines which index today billions of pages. However, they are generic services and they try to aim the largest number of users without considering their information needs in the search process. Moreover, users use generally few words to formulate their queries giving incomplete specifications of their information needs. So dealing this problem within Web context using traditional approaches is vain. This paper presents a novel particle swarm optimization approach for Web information retrieval. It uses relevance feedback to reformulate user query and thus improve the number of relevant results. In the authors' experimental results, they obtained a significant improvement of relevant results using their proposed approach comparing to what is obtained using only the user query into a search engine.