Concept Networks for Personalized Web Search Using Genetic Algorithm

Abstract

Web search engines gives users an initial point for their information hunt. The problem with the traditional search engine is that it retrieves the same set of web pages for all users even though each user has their own preference for a particular search. For retrieving web pages based on user's preference personalized search is needed. The main drawback of this method is that it cannot provide accurate result when user's preference changes or have to search something new. In the proposed method, a concept network is created to identify users search preferences. This concept network consists of list of related concepts based on the history of users’ previous search. This policy helps to retrieve web pages related to the context in which the user needs information. Genetic Algorithm (GA) is used when user searches for something new. GA is used to compare the user's concept network with other user's concept network for similarity, thus helps to get a better search in their area of interest.