Automatic generation of related work through summarizing citations

Jingqiang Chen*, Hai Zhuge

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

Related work is a component of a scientific paper, which introduces other researchers' relevant works and makes comparisons with the current author's work. Automatically generating the related work section of a writing paper provides a tool for researchers to accomplish the related work section efficiently without missing related works. This paper proposes an approach to automatically generating a related work section by comparing the main text of the paper being written with the citations of other papers that cite the same references. Our approach first collects the papers that cite the reference papers of the paper being written and extracts the corresponding citation sentences to form a citation document. It then extracts keywords from the citation document and the paper being written and constructs a graph of the keywords. Once the keywords that discriminate the two documents are determined, the minimum Steiner tree that covers the discriminative keywords and the topic keywords is generated. The summary is generated by extracting the sentences covering the Steiner tree. According to ROUGE evaluations, the experiments show that the citations are suitable for related work generation and our approach outperforms the three baseline methods of MEAD, LexRank, and ReWoS. This work verifies the general summarization method based on connotation and extension through citation.

Original languageEnglish
JournalConcurrency Computation
Early online date8 Sep 2017
DOIs
Publication statusE-pub ahead of print - 8 Sep 2017

Fingerprint

Citations
Steiner Tree
Experiments
Summarization
Baseline
Covering
Cover
Verify
Evaluation
Graph in graph theory
Experiment

Keywords

  • Citation
  • Related work generation
  • Summarization

Cite this

@article{c79428a85c4e452cbadbb8aed1c7456c,
title = "Automatic generation of related work through summarizing citations",
abstract = "Related work is a component of a scientific paper, which introduces other researchers' relevant works and makes comparisons with the current author's work. Automatically generating the related work section of a writing paper provides a tool for researchers to accomplish the related work section efficiently without missing related works. This paper proposes an approach to automatically generating a related work section by comparing the main text of the paper being written with the citations of other papers that cite the same references. Our approach first collects the papers that cite the reference papers of the paper being written and extracts the corresponding citation sentences to form a citation document. It then extracts keywords from the citation document and the paper being written and constructs a graph of the keywords. Once the keywords that discriminate the two documents are determined, the minimum Steiner tree that covers the discriminative keywords and the topic keywords is generated. The summary is generated by extracting the sentences covering the Steiner tree. According to ROUGE evaluations, the experiments show that the citations are suitable for related work generation and our approach outperforms the three baseline methods of MEAD, LexRank, and ReWoS. This work verifies the general summarization method based on connotation and extension through citation.",
keywords = "Citation, Related work generation, Summarization",
author = "Jingqiang Chen and Hai Zhuge",
year = "2017",
month = "9",
day = "8",
doi = "10.1002/cpe.4261",
language = "English",
journal = "Concurrency and Computation",
issn = "1532-0626",
publisher = "John Wiley and Sons Ltd",

}

Automatic generation of related work through summarizing citations. / Chen, Jingqiang; Zhuge, Hai.

In: Concurrency Computation, 08.09.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automatic generation of related work through summarizing citations

AU - Chen, Jingqiang

AU - Zhuge, Hai

PY - 2017/9/8

Y1 - 2017/9/8

N2 - Related work is a component of a scientific paper, which introduces other researchers' relevant works and makes comparisons with the current author's work. Automatically generating the related work section of a writing paper provides a tool for researchers to accomplish the related work section efficiently without missing related works. This paper proposes an approach to automatically generating a related work section by comparing the main text of the paper being written with the citations of other papers that cite the same references. Our approach first collects the papers that cite the reference papers of the paper being written and extracts the corresponding citation sentences to form a citation document. It then extracts keywords from the citation document and the paper being written and constructs a graph of the keywords. Once the keywords that discriminate the two documents are determined, the minimum Steiner tree that covers the discriminative keywords and the topic keywords is generated. The summary is generated by extracting the sentences covering the Steiner tree. According to ROUGE evaluations, the experiments show that the citations are suitable for related work generation and our approach outperforms the three baseline methods of MEAD, LexRank, and ReWoS. This work verifies the general summarization method based on connotation and extension through citation.

AB - Related work is a component of a scientific paper, which introduces other researchers' relevant works and makes comparisons with the current author's work. Automatically generating the related work section of a writing paper provides a tool for researchers to accomplish the related work section efficiently without missing related works. This paper proposes an approach to automatically generating a related work section by comparing the main text of the paper being written with the citations of other papers that cite the same references. Our approach first collects the papers that cite the reference papers of the paper being written and extracts the corresponding citation sentences to form a citation document. It then extracts keywords from the citation document and the paper being written and constructs a graph of the keywords. Once the keywords that discriminate the two documents are determined, the minimum Steiner tree that covers the discriminative keywords and the topic keywords is generated. The summary is generated by extracting the sentences covering the Steiner tree. According to ROUGE evaluations, the experiments show that the citations are suitable for related work generation and our approach outperforms the three baseline methods of MEAD, LexRank, and ReWoS. This work verifies the general summarization method based on connotation and extension through citation.

KW - Citation

KW - Related work generation

KW - Summarization

UR - http://www.scopus.com/inward/record.url?scp=85028982713&partnerID=8YFLogxK

U2 - 10.1002/cpe.4261

DO - 10.1002/cpe.4261

M3 - Article

AN - SCOPUS:85028982713

JO - Concurrency and Computation

JF - Concurrency and Computation

SN - 1532-0626

ER -