Home // CONTENT 2011, The Third International Conference on Creative Content Technologies // View article

Using Frame-based Lexical Chains for Extracting Key Points from Texts

Authors:
Sudabeh Mohamadi
Kambiz Badie
Ali Moeini

Keywords: automatic summarization; keyphrase extraction; abstract; lexical chain; generalization.

Abstract:
In last years, many automatic systems have been designed for text summarization and extracting key phrases, but still no system has been suggested for extracting key points. According to our definition, key points are high-level concepts extractable from a text that consist of words that may not necessarily exist in the original text. In this paper, we try to design an automatic system for extracting key points by using lexical chains. In this system, we use FrameNet for shallow semantic parsing of texts. As the final attempt, we present the set of tuples that contain important concepts of an original text with the related semantic roles. With use of generalization from parts onto whole, we can then have the claim of extracting a higher-level concept, which stands for a key point. Comparing the output of this system with human abstract, we perceived that 42 percent of cases generated by this system are similar to those generated by human.

Pages: 68 to 73

Copyright: Copyright (c) IARIA, 2011

Publication date: September 25, 2011

Published in: conference

ISSN: 2308-4162

ISBN: 978-1-61208-157-1

Location: Rome, Italy

Dates: from September 25, 2011 to September 30, 2011