Title: Summarising text with a genetic algorithm-based sentence extraction

Authors: Vahed Qazvinian, Leila Sharif Hassanabadi, Ramin Halavati

Addresses: Department of Computer Engineering, Sharif University of Technology, Tehran, Iran. ' Department of Computer Sciences, Shahid Beheshti University, Tehran, Iran. ' Department of Computer Engineering, Sharif University of Technology, Tehran, Iran

Abstract: Automatic text summarisation has long been studied and used. The growth in the amount of information on the web results in more demands for automatic methods for text summarisation. Designing a system to produce human-quality summaries is difficult and therefore, many researchers have focused on sentence or paragraph extraction, which is a kind of summarisation. In this paper, we introduce a new method to make such extracts. Genetic Algorithm (GA)-based sentence selection is used to make a summary, and once the summary is created, it is evaluated using a fitness function. The fitness function is based on three following factors: Readability Factor (RF), Cohesion Factor (CF) and Topic-Relation Factor (TRF). In this paper, we introduce these factors and discuss the Genetic Algorithm with the specific fitness function. Evaluation results are also shown and discussed in the paper.

Keywords: automatic text summarisation; sentence extraction; genetic algorithms; GAs; sentence selection; text summary.

DOI: 10.1504/IJKMS.2008.019750

International Journal of Knowledge Management Studies, 2008 Vol.2 No.4, pp.426 - 444

Published online: 29 Jul 2008 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article