Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization

Zhongyu Wei, Wei Gao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

16 Citations (Scopus)

Abstract

Single-document summarization is a challenging task. In this paper, we explore effective ways using the tweets linking to news for generating extractive summary of each document. We reveal the very basic value of tweets that can be utilized by regarding every tweet as a vote for candidate sentences. Base on such finding, we resort to unsupervised summarization models by leveraging the linking tweets to master the ranking of candidate extracts via random walk on a heterogeneous graph. The advantage is that we can use the linking tweets to opportunistically "supervise" the summarization with no need of reference summaries. Furthermore, we analyze the influence of the volume and latency of tweets on the quality of output summaries since tweets come after news release. Compared to truly supervised summarizer unaware of tweets, our method achieves significantly better results with reasonably small tradeoff on latency; compared to the same using tweets as auxiliary features, our method is comparable while needing less tweets and much shorter time to achieve significant outperformance.

Original languageEnglish
Title of host publicationSIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval
PublisherAssociation for Computing Machinery, Inc
Pages1003-1006
Number of pages4
ISBN (Print)9781450336215
DOIs
Publication statusPublished - 9 Aug 2015
Event38th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2015 - Santiago, Chile
Duration: 9 Aug 201513 Aug 2015

Other

Other38th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2015
CountryChile
CitySantiago
Period9/8/1513/8/15

Keywords

  • Highlights
  • Single-document summarization
  • Tweets

ASJC Scopus subject areas

  • Information Systems
  • Software

Cite this

Wei, Z., & Gao, W. (2015). Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization. In SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 1003-1006). Association for Computing Machinery, Inc. https://doi.org/10.1145/2766462.2767835

Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization. / Wei, Zhongyu; Gao, Wei.

SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2015. p. 1003-1006.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wei, Z & Gao, W 2015, Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization. in SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, pp. 1003-1006, 38th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2015, Santiago, Chile, 9/8/15. https://doi.org/10.1145/2766462.2767835
Wei Z, Gao W. Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization. In SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc. 2015. p. 1003-1006 https://doi.org/10.1145/2766462.2767835
Wei, Zhongyu ; Gao, Wei. / Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization. SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2015. pp. 1003-1006
@inproceedings{dc6a375f77e04f4faa47bab0965a4215,
title = "Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization",
abstract = "Single-document summarization is a challenging task. In this paper, we explore effective ways using the tweets linking to news for generating extractive summary of each document. We reveal the very basic value of tweets that can be utilized by regarding every tweet as a vote for candidate sentences. Base on such finding, we resort to unsupervised summarization models by leveraging the linking tweets to master the ranking of candidate extracts via random walk on a heterogeneous graph. The advantage is that we can use the linking tweets to opportunistically {"}supervise{"} the summarization with no need of reference summaries. Furthermore, we analyze the influence of the volume and latency of tweets on the quality of output summaries since tweets come after news release. Compared to truly supervised summarizer unaware of tweets, our method achieves significantly better results with reasonably small tradeoff on latency; compared to the same using tweets as auxiliary features, our method is comparable while needing less tweets and much shorter time to achieve significant outperformance.",
keywords = "Highlights, Single-document summarization, Tweets",
author = "Zhongyu Wei and Wei Gao",
year = "2015",
month = "8",
day = "9",
doi = "10.1145/2766462.2767835",
language = "English",
isbn = "9781450336215",
pages = "1003--1006",
booktitle = "SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Gibberish, assistant, or master? Using tweets linking to news for extractive single-document summarization

AU - Wei, Zhongyu

AU - Gao, Wei

PY - 2015/8/9

Y1 - 2015/8/9

N2 - Single-document summarization is a challenging task. In this paper, we explore effective ways using the tweets linking to news for generating extractive summary of each document. We reveal the very basic value of tweets that can be utilized by regarding every tweet as a vote for candidate sentences. Base on such finding, we resort to unsupervised summarization models by leveraging the linking tweets to master the ranking of candidate extracts via random walk on a heterogeneous graph. The advantage is that we can use the linking tweets to opportunistically "supervise" the summarization with no need of reference summaries. Furthermore, we analyze the influence of the volume and latency of tweets on the quality of output summaries since tweets come after news release. Compared to truly supervised summarizer unaware of tweets, our method achieves significantly better results with reasonably small tradeoff on latency; compared to the same using tweets as auxiliary features, our method is comparable while needing less tweets and much shorter time to achieve significant outperformance.

AB - Single-document summarization is a challenging task. In this paper, we explore effective ways using the tweets linking to news for generating extractive summary of each document. We reveal the very basic value of tweets that can be utilized by regarding every tweet as a vote for candidate sentences. Base on such finding, we resort to unsupervised summarization models by leveraging the linking tweets to master the ranking of candidate extracts via random walk on a heterogeneous graph. The advantage is that we can use the linking tweets to opportunistically "supervise" the summarization with no need of reference summaries. Furthermore, we analyze the influence of the volume and latency of tweets on the quality of output summaries since tweets come after news release. Compared to truly supervised summarizer unaware of tweets, our method achieves significantly better results with reasonably small tradeoff on latency; compared to the same using tweets as auxiliary features, our method is comparable while needing less tweets and much shorter time to achieve significant outperformance.

KW - Highlights

KW - Single-document summarization

KW - Tweets

UR - http://www.scopus.com/inward/record.url?scp=84953742789&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84953742789&partnerID=8YFLogxK

U2 - 10.1145/2766462.2767835

DO - 10.1145/2766462.2767835

M3 - Conference contribution

AN - SCOPUS:84953742789

SN - 9781450336215

SP - 1003

EP - 1006

BT - SIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval

PB - Association for Computing Machinery, Inc

ER -