A text parsing algorithm for discourse tree in the framework of rhetorical structure theory for indexing.

Muhammad Shoaib; Abad Ali Shah

No Detail Found!

A text parsing algorithm for discourse tree in the framework of rhetorical structure theory for indexing.

Author(s)

Muhammad Shoaib Department of Computer Science and Engineering, University of Engineering and Technology, Lahore, Pakistan

Abad Ali Shah Department of Computer Science and Engineering, University of Engineering and Technology, Lahore, Pakistan

Abstract

The problem of finding index terms using Rhetorical Structure theory involves segmentation of text, hypothesize the rhetorical relations that hold among textual units or segments and produce valid rhetorical structure trees for unrestricted natural language texts. Development of an algorithm for parsing the text in the framework of RST and representing it in the form of tree whose nodes contains relations and text spans is the main focus of this paper. The authors have assumed that a node that is closer to a nucleus node is a stronger one and others as weaker nodes. Initial assessment of weights has been done on this assumption. The output of this algorithm will be used for the development of RST based Indexer where weights will be assigned dynamically. The algorithm can be enhanced for handling other types of information like multimedia.

Publication Details

Page(s) 127-133

DOI DOI not available

Published Journal: Science International, Volume: 18, Issue: 2, Year: 2006

Request the full text from PASTIC