Measuring relevance with named entity based patterns in topic-focused document summarization

Furu Wei; Wenjie Li; Yanxiang He

doi:10.1109/NLPKE.2007.4368020

Measuring relevance with named entity based patterns in topic-focused document summarization

Furu Wei, Wenjie Li, Yanxiang He

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

In this paper, the role of named entity based patterns is emphasized in measuring the document sentences and topic relevance for topic-focused extractive summarization. Patterns are defined as the informative, semantic-sensitive text bi-grams consisting of at least one named entity or the semantic class of a named entity. They are extracted automatically according to eight pre-specified templates. Question types are also taken into consideration if they are available when dealing with topic questions. To alleviate problems with coverage, pattern and uni-gram models are integrated together to compensate each other in similarity calculation. Automatic ROUGE evaluations indicate that the proposed idea can produce a very good system that tops the best-performing system at Document Understanding Conference (DUC) 2005.

Original language	English
Title of host publication	IEEE NLP-KE 2007 - Proceedings of International Conference on Natural Language Processing and Knowledge Engineering
Pages	111-118
Number of pages	8
DOIs	https://doi.org/10.1109/NLPKE.2007.4368020
Publication status	Published - 1 Dec 2007
Event	International Conference on Natural Language Processing and Knowledge Engineering, IEEE NLP-KE 2007 - Beijing, China Duration: 30 Aug 2007 → 1 Sept 2007

Conference

Conference	International Conference on Natural Language Processing and Knowledge Engineering, IEEE NLP-KE 2007
Country/Territory	China
City	Beijing
Period	30/08/07 → 1/09/07

ASJC Scopus subject areas

Computer Science Applications
Information Systems
Information Systems and Management

More information

10.1109/NLPKE.2007.4368020

Cite this

@inproceedings{11301d5343344c6bad745f50bd26ad4d,

title = "Measuring relevance with named entity based patterns in topic-focused document summarization",

abstract = "In this paper, the role of named entity based patterns is emphasized in measuring the document sentences and topic relevance for topic-focused extractive summarization. Patterns are defined as the informative, semantic-sensitive text bi-grams consisting of at least one named entity or the semantic class of a named entity. They are extracted automatically according to eight pre-specified templates. Question types are also taken into consideration if they are available when dealing with topic questions. To alleviate problems with coverage, pattern and uni-gram models are integrated together to compensate each other in similarity calculation. Automatic ROUGE evaluations indicate that the proposed idea can produce a very good system that tops the best-performing system at Document Understanding Conference (DUC) 2005.",

author = "Furu Wei and Wenjie Li and Yanxiang He",

year = "2007",

month = dec,

day = "1",

doi = "10.1109/NLPKE.2007.4368020",

language = "English",

isbn = "9781424416103",

pages = "111--118",

booktitle = "IEEE NLP-KE 2007 - Proceedings of International Conference on Natural Language Processing and Knowledge Engineering",

note = "International Conference on Natural Language Processing and Knowledge Engineering, IEEE NLP-KE 2007 ; Conference date: 30-08-2007 Through 01-09-2007",

}

Wei, F, Li, W & He, Y 2007, Measuring relevance with named entity based patterns in topic-focused document summarization. in IEEE NLP-KE 2007 - Proceedings of International Conference on Natural Language Processing and Knowledge Engineering., 4368020, pp. 111-118, International Conference on Natural Language Processing and Knowledge Engineering, IEEE NLP-KE 2007, Beijing, China, 30/08/07. https://doi.org/10.1109/NLPKE.2007.4368020

Measuring relevance with named entity based patterns in topic-focused document summarization. / Wei, Furu; Li, Wenjie; He, Yanxiang.
IEEE NLP-KE 2007 - Proceedings of International Conference on Natural Language Processing and Knowledge Engineering. 2007. p. 111-118 4368020.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Measuring relevance with named entity based patterns in topic-focused document summarization

AU - Wei, Furu

AU - Li, Wenjie

AU - He, Yanxiang

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In this paper, the role of named entity based patterns is emphasized in measuring the document sentences and topic relevance for topic-focused extractive summarization. Patterns are defined as the informative, semantic-sensitive text bi-grams consisting of at least one named entity or the semantic class of a named entity. They are extracted automatically according to eight pre-specified templates. Question types are also taken into consideration if they are available when dealing with topic questions. To alleviate problems with coverage, pattern and uni-gram models are integrated together to compensate each other in similarity calculation. Automatic ROUGE evaluations indicate that the proposed idea can produce a very good system that tops the best-performing system at Document Understanding Conference (DUC) 2005.

AB - In this paper, the role of named entity based patterns is emphasized in measuring the document sentences and topic relevance for topic-focused extractive summarization. Patterns are defined as the informative, semantic-sensitive text bi-grams consisting of at least one named entity or the semantic class of a named entity. They are extracted automatically according to eight pre-specified templates. Question types are also taken into consideration if they are available when dealing with topic questions. To alleviate problems with coverage, pattern and uni-gram models are integrated together to compensate each other in similarity calculation. Automatic ROUGE evaluations indicate that the proposed idea can produce a very good system that tops the best-performing system at Document Understanding Conference (DUC) 2005.

UR - http://www.scopus.com/inward/record.url?scp=47749152958&partnerID=8YFLogxK

U2 - 10.1109/NLPKE.2007.4368020

DO - 10.1109/NLPKE.2007.4368020

M3 - Conference article published in proceeding or book

SN - 9781424416103

SP - 111

EP - 118

BT - IEEE NLP-KE 2007 - Proceedings of International Conference on Natural Language Processing and Knowledge Engineering

T2 - International Conference on Natural Language Processing and Knowledge Engineering, IEEE NLP-KE 2007

Y2 - 30 August 2007 through 1 September 2007

ER -

Measuring relevance with named entity based patterns in topic-focused document summarization

Abstract

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this