Metasearch Via the Co-Citation Graph

—In spite of numerous search engines available on the web, no single engine is capable of performing “better” under all circumstances. This being the case, metasearch engines have appeared. One major issue in metasearch engine design is the merging problem: given the ranked lists of pages returned by multiple search engines in response to a given query, what is the best way to combine these lists into a single list? In this paper, we present an algorithm based on the co-citation graph constructed from search engines and returned lists of pages to handle this issue. Our algorithm can be implemented in a simple and naive manner while based on a rigorous and intuitive treatment of the problem. Unlike those techniques that rely on the relevant scores of pages, our algorithm can be implemented using only the rank information. Terms Algorithms, performance, experimentation Keywords Merging, reranking, meta search

• Type: Conference
• Year: 2003
• Where: IC
• Authors: Hyun-Chul Lee
• Keywords: Engines, IC 2003, IC 2007, Metasearch Engine, Search Engines