An unsupervised Bayesian modelling approach to storyline detection from news articles

Deyu Zhou, Haiyang Xu, Yulan He

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

Abstract

Storyline detection from news articles aims at summarizing events described under a certain news topic and revealing how those events evolve over time. It is a difficult task because it requires first the detection of events from news articles published in different time periods and then the construction of storylines by linking events into coherent news stories. Moreover, each storyline has different hierarchical structures which are dependent across epochs. Existing approaches often ignore the dependency of hierarchical structures in storyline generation. In this paper, we propose an unsupervised Bayesian model, called dynamic storyline detection model, to extract structured representations and evolution patterns of storylines. The proposed model is evaluated on a large scale news corpus. Experimental results show that our proposed model outperforms several baseline approaches.

Original languageEnglish
Title of host publicationEMNLP 2015 : Conference on empirical methods in natural language processing
Subtitle of host publicationproceedings of the Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015)
EditorsRobert Berwick, Anna Korhonen, Alessandro Lenci, et al
Place of PublicationRed Hook, NY (US)
PublisherAssociation for Computational Linguistics
Pages1943-1948
Number of pages6
ISBN (Print)978-1-941643-32-7
Publication statusPublished - 2015
Event6th Workshop on Cognitive Aspects of Computational Language Learning - Lisbon, Portugal
Duration: 17 Sep 201521 Sep 2015

Workshop

Workshop6th Workshop on Cognitive Aspects of Computational Language Learning
CountryPortugal
CityLisbon
Period17/09/1521/09/15
OtherWorkshop as past of the Conference on Empirical Methods in Natural Language Processing

Fingerprint

Dynamic models

Bibliographical note

Conference on Empirical Methods in Natural Language Processing: Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015)

Cite this

Zhou, D., Xu, H., & He, Y. (2015). An unsupervised Bayesian modelling approach to storyline detection from news articles. In R. Berwick, A. Korhonen, A. Lenci, & et al (Eds.), EMNLP 2015 : Conference on empirical methods in natural language processing: proceedings of the Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015) (pp. 1943-1948). Red Hook, NY (US): Association for Computational Linguistics.
Zhou, Deyu ; Xu, Haiyang ; He, Yulan. / An unsupervised Bayesian modelling approach to storyline detection from news articles. EMNLP 2015 : Conference on empirical methods in natural language processing: proceedings of the Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015). editor / Robert Berwick ; Anna Korhonen ; Alessandro Lenci ; et al. Red Hook, NY (US) : Association for Computational Linguistics, 2015. pp. 1943-1948
@inproceedings{569c85e5bb4f495dac19c649f03db7e7,
title = "An unsupervised Bayesian modelling approach to storyline detection from news articles",
abstract = "Storyline detection from news articles aims at summarizing events described under a certain news topic and revealing how those events evolve over time. It is a difficult task because it requires first the detection of events from news articles published in different time periods and then the construction of storylines by linking events into coherent news stories. Moreover, each storyline has different hierarchical structures which are dependent across epochs. Existing approaches often ignore the dependency of hierarchical structures in storyline generation. In this paper, we propose an unsupervised Bayesian model, called dynamic storyline detection model, to extract structured representations and evolution patterns of storylines. The proposed model is evaluated on a large scale news corpus. Experimental results show that our proposed model outperforms several baseline approaches.",
author = "Deyu Zhou and Haiyang Xu and Yulan He",
note = "Conference on Empirical Methods in Natural Language Processing: Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015)",
year = "2015",
language = "English",
isbn = "978-1-941643-32-7",
pages = "1943--1948",
editor = "Robert Berwick and Anna Korhonen and Alessandro Lenci and {et al}",
booktitle = "EMNLP 2015 : Conference on empirical methods in natural language processing",
publisher = "Association for Computational Linguistics",

}

Zhou, D, Xu, H & He, Y 2015, An unsupervised Bayesian modelling approach to storyline detection from news articles. in R Berwick, A Korhonen, A Lenci & et al (eds), EMNLP 2015 : Conference on empirical methods in natural language processing: proceedings of the Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015). Association for Computational Linguistics, Red Hook, NY (US), pp. 1943-1948, 6th Workshop on Cognitive Aspects of Computational Language Learning, Lisbon, Portugal, 17/09/15.

An unsupervised Bayesian modelling approach to storyline detection from news articles. / Zhou, Deyu; Xu, Haiyang; He, Yulan.

EMNLP 2015 : Conference on empirical methods in natural language processing: proceedings of the Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015). ed. / Robert Berwick; Anna Korhonen; Alessandro Lenci; et al. Red Hook, NY (US) : Association for Computational Linguistics, 2015. p. 1943-1948.

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

TY - GEN

T1 - An unsupervised Bayesian modelling approach to storyline detection from news articles

AU - Zhou, Deyu

AU - Xu, Haiyang

AU - He, Yulan

N1 - Conference on Empirical Methods in Natural Language Processing: Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015)

PY - 2015

Y1 - 2015

N2 - Storyline detection from news articles aims at summarizing events described under a certain news topic and revealing how those events evolve over time. It is a difficult task because it requires first the detection of events from news articles published in different time periods and then the construction of storylines by linking events into coherent news stories. Moreover, each storyline has different hierarchical structures which are dependent across epochs. Existing approaches often ignore the dependency of hierarchical structures in storyline generation. In this paper, we propose an unsupervised Bayesian model, called dynamic storyline detection model, to extract structured representations and evolution patterns of storylines. The proposed model is evaluated on a large scale news corpus. Experimental results show that our proposed model outperforms several baseline approaches.

AB - Storyline detection from news articles aims at summarizing events described under a certain news topic and revealing how those events evolve over time. It is a difficult task because it requires first the detection of events from news articles published in different time periods and then the construction of storylines by linking events into coherent news stories. Moreover, each storyline has different hierarchical structures which are dependent across epochs. Existing approaches often ignore the dependency of hierarchical structures in storyline generation. In this paper, we propose an unsupervised Bayesian model, called dynamic storyline detection model, to extract structured representations and evolution patterns of storylines. The proposed model is evaluated on a large scale news corpus. Experimental results show that our proposed model outperforms several baseline approaches.

UR - http://www.scopus.com/inward/record.url?scp=84959903033&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84959903033

SN - 978-1-941643-32-7

SP - 1943

EP - 1948

BT - EMNLP 2015 : Conference on empirical methods in natural language processing

A2 - Berwick, Robert

A2 - Korhonen, Anna

A2 - Lenci, Alessandro

A2 - et al, null

PB - Association for Computational Linguistics

CY - Red Hook, NY (US)

ER -

Zhou D, Xu H, He Y. An unsupervised Bayesian modelling approach to storyline detection from news articles. In Berwick R, Korhonen A, Lenci A, et al, editors, EMNLP 2015 : Conference on empirical methods in natural language processing: proceedings of the Sixth Workshop on Cognitive Aspects of Computational Language Learning (CogACLL-2015). Red Hook, NY (US): Association for Computational Linguistics. 2015. p. 1943-1948