Link Analysis Based on Rhetorical Relations for Multi-Document Summarization

Nik Adilah Hanin BINTI ZAHRI; Fumiyo FUKUMOTO; Suguru MATSUYOSHI

doi:10.1587/transinf.E96.D.1182

Link Analysis Based on Rhetorical Relations for Multi-Document Summarization

Nik Adilah Hanin BINTI ZAHRI, Fumiyo FUKUMOTO, Suguru MATSUYOSHI

Full Text Views

0

Share
Cite this

Summary :

This paper presents link analysis based on rhetorical relations with the aim of performing extractive summarization for multiple documents. We first extracted sentences with salient terms from individual document using statistical model. We then ranked the extracted sentences by measuring their relative importance according to their connectivity among the sentences in the document set using PageRank based on the rhetorical relations. The rhetorical relations were examined beforehand to determine which relations are crucial to this task, and the relations among sentences from documents were automatically identified by SVMs. We used the relations to emphasize important sentences during sentence ranking by PageRank and eliminate redundancy from the summary candidates. Our framework omits fully annotated sentences by humans and the evaluation results show that the combination of PageRank along with rhetorical relations does help to improve the quality of extractive summarization.

Publication: IEICE TRANSACTIONS on Information Vol.E96-D No.5 pp.1182-1191

Publication Date: 2013/05/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E96.D.1182

Type of Manuscript: PAPER

Category: Natural Language Processing

Cite this

Copy

Nik Adilah Hanin BINTI ZAHRI, Fumiyo FUKUMOTO, Suguru MATSUYOSHI, "Link Analysis Based on Rhetorical Relations for Multi-Document Summarization" in IEICE TRANSACTIONS on Information, vol. E96-D, no. 5, pp. 1182-1191, May 2013, doi: 10.1587/transinf.E96.D.1182.
Abstract: This paper presents link analysis based on rhetorical relations with the aim of performing extractive summarization for multiple documents. We first extracted sentences with salient terms from individual document using statistical model. We then ranked the extracted sentences by measuring their relative importance according to their connectivity among the sentences in the document set using PageRank based on the rhetorical relations. The rhetorical relations were examined beforehand to determine which relations are crucial to this task, and the relations among sentences from documents were automatically identified by SVMs. We used the relations to emphasize important sentences during sentence ranking by PageRank and eliminate redundancy from the summary candidates. Our framework omits fully annotated sentences by humans and the evaluation results show that the combination of PageRank along with rhetorical relations does help to improve the quality of extractive summarization.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E96.D.1182/_p

Copy

@ARTICLE{e96-d_5_1182,
author={Nik Adilah Hanin BINTI ZAHRI, Fumiyo FUKUMOTO, Suguru MATSUYOSHI, },
journal={IEICE TRANSACTIONS on Information},
title={Link Analysis Based on Rhetorical Relations for Multi-Document Summarization},
year={2013},
volume={E96-D},
number={5},
pages={1182-1191},
abstract={This paper presents link analysis based on rhetorical relations with the aim of performing extractive summarization for multiple documents. We first extracted sentences with salient terms from individual document using statistical model. We then ranked the extracted sentences by measuring their relative importance according to their connectivity among the sentences in the document set using PageRank based on the rhetorical relations. The rhetorical relations were examined beforehand to determine which relations are crucial to this task, and the relations among sentences from documents were automatically identified by SVMs. We used the relations to emphasize important sentences during sentence ranking by PageRank and eliminate redundancy from the summary candidates. Our framework omits fully annotated sentences by humans and the evaluation results show that the combination of PageRank along with rhetorical relations does help to improve the quality of extractive summarization.},
keywords={},
doi={10.1587/transinf.E96.D.1182},
ISSN={1745-1361},
month={May},}

Copy

TY - JOUR
TI - Link Analysis Based on Rhetorical Relations for Multi-Document Summarization
T2 - IEICE TRANSACTIONS on Information
SP - 1182
EP - 1191
AU - Nik Adilah Hanin BINTI ZAHRI
AU - Fumiyo FUKUMOTO
AU - Suguru MATSUYOSHI
PY - 2013
DO - 10.1587/transinf.E96.D.1182
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2013
AB - This paper presents link analysis based on rhetorical relations with the aim of performing extractive summarization for multiple documents. We first extracted sentences with salient terms from individual document using statistical model. We then ranked the extracted sentences by measuring their relative importance according to their connectivity among the sentences in the document set using PageRank based on the rhetorical relations. The rhetorical relations were examined beforehand to determine which relations are crucial to this task, and the relations among sentences from documents were automatically identified by SVMs. We used the relations to emphasize important sentences during sentence ranking by PageRank and eliminate redundancy from the summary candidates. Our framework omits fully annotated sentences by humans and the evaluation results show that the combination of PageRank along with rhetorical relations does help to improve the quality of extractive summarization.
ER -