Skip to content

Proceedings - Real-time Summarization 2018

Overview of the TREC 2018 Real-Time Summarization Track

Royal Sequiera, Luchen Tan, Jimmy Lin

Abstract

The TREC 2018 Real-Time Summarization (RTS) Track is the third iteration of a community effort to explore techniques, algorithms, and systems that automatically monitor streams of social media posts such as tweets on Twi‹er to address users' prospective information needs. These needs are articulated as “interest profiles”, akin to topics in ad hoc retrieval. In our formulation of real-time summarization, the goal is for a system to deliver relevant and novel content to users in a timely fashion. We refer to these messages generically as “updates”. As with previous iterations of the evaluation, the task setup required participating systems to monitor the live Twi‹er sample stream during a pre-defined evaluation period, this year beginning Monday July 23, 2018 00:00:00 UTC and ending Friday August 3, 2018 23:59:59 UTC. The interest profiles were distributed to participants ahead of time. [...]

Bibtex 
@inproceedings{DBLP:conf/trec/SequieraTL18,
    author = {Royal Sequiera and Luchen Tan and Jimmy Lin},
    editor = {Ellen M. Voorhees and Angela Ellis},
    title = {Overview of the {TREC} 2018 Real-Time Summarization Track},
    booktitle = {Proceedings of the Twenty-Seventh Text REtrieval Conference, {TREC} 2018, Gaithersburg, Maryland, USA, November 14-16, 2018},
    series = {{NIST} Special Publication},
    volume = {500-331},
    publisher = {National Institute of Standards and Technology {(NIST)}},
    year = {2018},
    url = {https://trec.nist.gov/pubs/trec27/papers/Overview-RTS.pdf},
    timestamp = {Wed, 03 Feb 2021 00:00:00 +0100},
    biburl = {https://dblp.org/rec/conf/trec/SequieraTL18.bib},
    bibsource = {dblp computer science bibliography, https://dblp.org}
}

IRIT at TREC Real-Time Summarization 2018

Abdelhamid Chellal, Mohand Boughanem

Abstract

This paper presents the participation of the IRIT laboratory (University of Toulouse) to the Real-Time Summarization track of TREC RTS 2018. This track is consisting of two scenarios ( A: push notification and B: Email digest) which tackle the challenge of fulfilling the prospective and the retrospection information needs repressively. We submitted three runs for both scenarios A and B. For scenario A, we propose to use a supervised learning approach to build a binary classifier that predicts the relevance of an incoming tweet with respect to the topic of interest. The proposed approach leverages social signals as well as query dependent features to enhance the detection of relevant tweets. Additionally, we investigate the impact of the use of live relevance feedback to re-train the classier each time new relevance assessments are made available. For scenario B, the daily digest is generated by iteratively selecting the top tweets that pass the relevance filter with discarding the redundant ones.

Bibtex 
@inproceedings{DBLP:conf/trec/ChellalB18,
    author = {Abdelhamid Chellal and Mohand Boughanem},
    editor = {Ellen M. Voorhees and Angela Ellis},
    title = {{IRIT} at {TREC} Real-Time Summarization 2018},
    booktitle = {Proceedings of the Twenty-Seventh Text REtrieval Conference, {TREC} 2018, Gaithersburg, Maryland, USA, November 14-16, 2018},
    series = {{NIST} Special Publication},
    volume = {500-331},
    publisher = {National Institute of Standards and Technology {(NIST)}},
    year = {2018},
    url = {https://trec.nist.gov/pubs/trec27/papers/IRIT-RT.pdf},
    timestamp = {Thu, 12 Mar 2020 00:00:00 +0100},
    biburl = {https://dblp.org/rec/conf/trec/ChellalB18.bib},
    bibsource = {dblp computer science bibliography, https://dblp.org}
}

GPLSI at TREC 2018 RTS Track

Javi Fernández, Fernando Llopis, Yoan Gutiérrez, Patricio Martínez-Barco, José M. Gómez, Rafael Muñoz

Abstract

n this paper we present our contribution for the TREC 2018 Real-Time Summarization track. This task contains two scenarios: push notifications, and email digest. We participated in both, submitting three runs on each one. Our main goal was to evaluate the effectiveness the techniques employed in Social Analytics, a reputation analysis platform, which finds relevant tweets for specific topics. Here, we describe these techniques, and discuss the results obtained

Bibtex 
@inproceedings{DBLP:conf/trec/FernandezLGMG018,
    author = {Javi Fern{\'{a}}ndez and Fernando Llopis and Yoan Guti{\'{e}}rrez and Patricio Mart{\'{\i}}nez{-}Barco and Jos{\'{e}} M. G{\'{o}}mez and Rafael Mu{\~{n}}oz},
    editor = {Ellen M. Voorhees and Angela Ellis},
    title = {{GPLSI} at {TREC} 2018 {RTS} Track},
    booktitle = {Proceedings of the Twenty-Seventh Text REtrieval Conference, {TREC} 2018, Gaithersburg, Maryland, USA, November 14-16, 2018},
    series = {{NIST} Special Publication},
    volume = {500-331},
    publisher = {National Institute of Standards and Technology {(NIST)}},
    year = {2018},
    url = {https://trec.nist.gov/pubs/trec27/papers/UA-GPLSI-RT.pdf},
    timestamp = {Thu, 12 Mar 2020 00:00:00 +0100},
    biburl = {https://dblp.org/rec/conf/trec/FernandezLGMG018.bib},
    bibsource = {dblp computer science bibliography, https://dblp.org}
}