Predicting relevant news events for timeline summaries

Tran, Giang Binh; Alrifai, Mohammad; Nguyen, Dat Quoc

doi:10.1145/2487788.2487829

Cited by 37 publications

(49 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prior approaches dedicated specifically to date selection are Tran et al (2013a) and Kessler et al (2012). 3 They use supervised machine learning methods that score dates independently of each other.…”

Section: Related Workmentioning

confidence: 99%

“…Therefore, timelines of variable length can be constructed. Like (Kessler et al, 2012;Tran et al, 2013a), we also assume that we have a corpus C, consisting of news articles about the main event. This corpus gives evidence about the dates in [t 1 , t 2 ].…”

Section: Problem Definitionmentioning

confidence: 99%

“…The first setting is the one used in previous work such as (Kessler et al, 2012;Tran et al, 2013a). It is also the only one that can be used if only one timeline per event is considered as in Kessler et al (2012).…”

Section: Experimental Settingsmentioning

confidence: 99%

“…For example, the creator of the startup Timeline says that it initially took a multiperson team a full work day to create a single timeline. 1 Therefore, automatic timeline summarization (TLS) has emerged as an NLP task in the past few years (Tran et al, 2013a;Kessler et al, 2012;Nguyen et al, 2014;Yan et al, 2011b;Yan et al, 2011a;Wang et al, 2012;Tran et al, 2013b;Tran et al, 2015). TLS has been divided into two subtasks: (i) ranking the dates between beginning and end of the timeline in order of importance, to achieve date selection and (ii) generating a good daily summary for each of the selected dates.…”

Section: Introductionmentioning

confidence: 99%

“…Date selection may be partially subjective: different journalists might include different dates. 2 Existing approaches to date selection (Kessler et al, 2012;Tran et al, 2013a) use supervised machine learning, where each date receives a score for ranking the dates. Features used (such as frequency of date mention) are extracted from a corpus of event-related newspaper articles.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Joint Graphical Models for Date Selection in Timeline Summarization

Tran

Herder

Markert

2015

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Confere

View full text Add to dashboard Cite

Automatic timeline summarization (TLS) generates precise, dated overviews over (often prolonged) events, such as wars or economic crises. One subtask of TLS selects the most important dates for an event within a certain time frame. Date selection has up to now been handled via supervised machine learning approaches that estimate the importance of each date separately, using features such as the frequency of date mentions in news corpora. This approach neglects interactions between different dates that occur due to connections between subevents. We therefore suggest a joint graphical model for date selection. Even unsupervised versions of this model perform as well as supervised state-of-theart approaches. With parameter tuning on training data, it outperforms prior supervised models by a considerable margin.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Problem Definitionmentioning

confidence: 99%

Section: Experimental Settingsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Joint Graphical Models for Date Selection in Timeline Summarization

Tran

Herder

Markert

2015

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Confere

View full text Add to dashboard Cite

show abstract

Timeline Summarization from Relevant Headlines

Tran

Alrifai

Herder

2015

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Timeline summaries are an effective way for helping newspaper readers to keep track of long-lasting news stories, such as the Egypt revolution. A good timeline summary provides a concise description of only the main events, while maintaining good understandability. As manual construction of timelines is very time-consuming, there is a need for automatic approaches. However, automatic selection of relevant events is challenging due to the large amount of news articles published every day. Furthermore, current state-of-the-art systems produce summaries that are suboptimal in terms of relevance and understandability. We present a new approach that exploits the headlines of online news articles instead of the articles' full text. The quantitative and qualitative results from our user studies confirm that our method outperforms state-of-the-art system in these aspects.

show abstract

Graphical Models with Content Relevance for Crucial Date Detection in Social Media Event

Ding

2018

Knowledge Science, Engineering and Management

View full text Add to dashboard Cite

Predicting relevant news events for timeline summaries

Cited by 37 publications

References 5 publications

Joint Graphical Models for Date Selection in Timeline Summarization

Joint Graphical Models for Date Selection in Timeline Summarization

Timeline Summarization from Relevant Headlines

Graphical Models with Content Relevance for Crucial Date Detection in Social Media Event

Contact Info

Product

Resources

About