The Research of Information Retrieval Technology Based on Semantic Analysis

Abstract:

The common information retrieval technology is mainly based on keyword matching and this kind of method only focuse on the optimization algorithm, ignoring the semantic research. This does not solve the fundamental semantic multiplicity, retrieve diversity, related web undetected, sort unstandardized. This paper is a study of these problems arise for the current proposed MIRSA information retrieval model based on semantic analysis. This model consists of the following four main key points: disambiguation method, semantic expansion algorithm, the search terms match strategy, web sorting algorithms. This model can effectively solve the problem of semantic multiplicity, avoid missed relevant pages and reasonably improve the sor of related pages.

• Periodical: Advanced Materials Research (Volumes 926-930)
• Pages: 2160-2163
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.926-930.2160
• Online since: May 2014
• Authors: Ming Xu*, Yun Ke
• Keywords: Information Retrieval, Semantic Analysis, Semantic Similarity

References

[1] Brin S, Page L: Computer Networks and ISDN Systems, Vo1. 1 (1998) No. 3, pp.107-117.

[2] Kleinberg J: Extended Versionin Journal of the ACM, Vo1. 1 (1999) No. 1, pp.47-51.

[3] Sun Jiantao, Zeng Huajun, Liu Huan: Proceedings of International World Wide Web, Vo1. 3 (2005) No. 1, pp.277-281.

[4] Maeve O'Brien, P. Mark, T. Keane: Principles andMethods. Data and Knowledge Engineering, Vol. 1 (1998) No. 6, pp.43-50.

[5] Nunda Stojanovic: Journal of China Agricultural University, Vo1. 6 (2003) No. 2, pp.3-10.