Automatic Text Summarization as a Text Extraction Strategy for Effective Automated Highlighting

Keywords
Thumbnail Image
Date
2018-02-26
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Automatic text highlighting is capable of becoming a new tool in textual information processing. Preliminary research is done to examine the potential of a new application for text summarization algorithms. A framework is built for parsing and highlighting PDF files, to which extraction strategies can be applied. Also a small dataset of highlighted documents is gathered to test the highlighting capabilities of four text summarization algorithms on. Thereby, ROUGE-N scores are obtained, the performance per task and the performance per algorithm are evaluated. In the end, the algorithms perform surprisingly well and may encourage more investment in automatic text highlighting.
Description
Citation
Faculty
Faculteit der Sociale Wetenschappen