Designing a Meta-Search Engine Considering the User’s Field of Interest

Existing meta-search engines return web search results based on the page relevancy to the query, their popularity and content. In addition, they disregard the user’s preferences or field of interest. It is necessary to provide a meta-search engine capable of ranking results considering the user’s field of interest. Social Networks can be useful to find the users’ tendencies, favorites, skills and interests. In this paper we propose MSE, a Meta-Search Engine for document retrieval utilizing social information of the user. In this approach, each user is associated with a user profile that captures his interests available from a social network he or she belongs to. The MSE receives search results from three underlying search engines. It extracts main phrases from the title and short description of each result. Then it clusters the main phrases by self-organizing feature map algorithm. Generated clusters are then ranked on the basis of the user profile. The more similar cluster label to the user’s field of interest gets the higher rank. We have compared the proposed MSE against two other meta-search engines. The experimental results show the efficiency and effectiveness of the proposed method

Keywords:

Clustering , Meta-Search Engine , Ranking , Search Relevance , Social Information