Research Article Open Access

A New Indexing Technique for Information Retrieval Systems Using Rhetorical Structure Theory (RST)

Muhammad Shoaib and Abad Ali Shah

Abstract

Effective information retrieval requires an efficient indexing technique. With the availability of huge volume of information, it has become necessary to capture the semantic of the document, which is almost impossible with the existing techniques. Moreover in the existing technique the weights once assign are remains unchangeable through out the cycle. In this paper an indexing technique assignment using Rhetorical Structure Theory with dynamic weight assignment technique has been presented. The nodes of Rhetorical Parsing tree contain relations and text spans which can be used for indexing by indexer. The results are promising for different texts. Enhancing the technique of NLP can improve the proposed algorithm to accommodate more relations and huge documents.

Journal of Computer Science
Volume 2 No. 3, 2006, 224-228

DOI: https://doi.org/10.3844/jcssp.2006.224.228

Submitted On: 25 October 2005 Published On: 31 March 2006

How to Cite: Shoaib, M. & Shah, A. A. (2006). A New Indexing Technique for Information Retrieval Systems Using Rhetorical Structure Theory (RST). Journal of Computer Science, 2(3), 224-228. https://doi.org/10.3844/jcssp.2006.224.228

  • 2,774 Views
  • 2,831 Downloads
  • 1 Citations

Download

Keywords

  • Indexing
  • RST
  • dynamic weight assignment