Multi-Agent Based Web Search with Heterogeneous Semantics

  • Authors:

  • Rui Huang;Zhongzhi Shi

  • Affiliations:

  • Key Laboratory of Intelligent Information Processing Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China 100190 and Graduate School of the Chinese Academy of Sciences, B ...;Key Laboratory of Intelligent Information Processing Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China 100190

  • Venue:
  • Agent Computing and Multi-Agent Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Relevance ranking is key to Web search in determining how results are retrieved and ordered. As keyword-based search does not guarantee relevance in meanings, semantic search has attracted enormous and growing interest to improve the accuracy of relevance ranking. Recently heterogeneous semantic information such as thesauruses, semantic markups and social annotations have been adopted in search respectively for this purpose. However, although to integrate more semantics would logically generate better search results in respect of semantic relevance, such integrated semantic search mechanism is still in absence and to be researched. This paper proposes a multi-agent based semantic search approach to integrate both keywords and heterogeneous semantics. Such integration is achieved through semantic query expansion, meta search of expanded queries in varieties of existing search engines, and aggregation of all search results at the semantic level. With respect to the great volumes of distributed and dynamic Web information, this multi-agent based approach not only guarantees efficiency and reliability of search, but also enables automatic and effective cooperations for semantic integration. Experiments show that the proposed approach can effectively integrate both keywords and heterogeneous semantics for Web search.