Generalized vector spaces model in information retrieval

Abstract

In information retrieval, it is common to model index terms and documents as vectore in a suitably defined vector space. The main di]ficulty with this approach is that the explicit repreeentation of term vectors is not known a priorL For th~ mason, the vector space model adopted by Salton for the SMART system treats the terms as a set of orthogonal vectom In such a model it is often necessary to adopt a separate, corrective procedure to take into account the correlations between terms. In this paper, we propose a systematic method (the generalized vector space model) to compute term correlations directly from automatic indexing scheme. We also demonstrate how such correlations can be included with minimal modification in the existing vector based information retrieval systems. The preliminary experimental . results obtained from the new model are very encouraging.

Word Sense Disambiguation for Exploiting Hierarchical Thesauri in Text Classification

“…The Semantic Smoothing Kernels have similar semantics to the GVSM model defined in [8]. A kernel definition based on the GVSM model is given below.…”

“…Experiments, using the WordNet HT, demonstrate that our WSD algorithm can be configured to exhibit very high precision, and thus can be considered appropriate for classification. In order to exploit the semantic relations inherent in the HT, we define a semantic kernel based on the general concept of GVSM kernels [8]. Finally, we have conducted experiments utilizing various sizes of training sets for the two largest Reuters-21578 categories and a corpus constructed from crawling editorial reviews of books from the Amazon website.…”

Intelligence Analysis Modeling

“…Vector space model lacks in that it treats each index term as a separate coordinate and assumes the terms as being orthogonal, which is deemed contrary to the reality where term relationships exist and index terms are not assigned independently of each other [18].…”

Building an Integrated Pan-European News Distribution Network

“…The most relevant semantically related business news applied to the current news item in focus are retrieved and provided via the network service. For this purpose the practical prototype as a proof of concept for the technical architecture and the semantic coupling of contents uses a search mechanism common in modern information retrieval systems: the Vector Space Model [Wong, 1985]. This approach is mainly used for search engines, based on natural language.…”

“…Research fields such as network communication [Yu, 2004;Birman, 2003] have been involved to create a scaleable peer-to-peer architecture, artificial intelligence is utilized to identify the most relevant related articles within the entire multinational network that can add to the quality of the business news currently in focus. The vector space model [Salton, 1983;Wong, 1985] has been utilized to provide easy access to related and most relevant business news articles within a multilingual and multinational context. This paper is focused on the integration of modern Internet technologies such as Web Services [Haas, 2002] and peer to peer architectures [Birman, 2003] to create a scalable and high traffic information exchange and distribution network.…”