Distant supervision for tweet classification using youtube labels

Walid Magdy, Hassan Sajjad, Tarek El-Ganainy, Fabrizio Sebastiani

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

6 Citations (Scopus)

Abstract

We study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another. In particular, we apply classes assigned to YouTube videos to tweet slinking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The experiments we have runs how that a tweet classifier trained via these automatically labelled data substantially outperforms an analogous classifier trained with a limited amount of manually labelled data.

Original languageEnglish
Title of host publicationProceedings of the 9th International Conference on Web and Social Media, ICWSM 2015
PublisherAAAI Press
Pages638-641
Number of pages4
ISBN (Print)9781577357339
Publication statusPublished - 2015
Event9th International Conference on Web and Social Media, ICWSM 2015 - Oxford, United Kingdom
Duration: 26 May 201529 May 2015

Other

Other9th International Conference on Web and Social Media, ICWSM 2015
CountryUnited Kingdom
CityOxford
Period26/5/1529/5/15

Fingerprint

Labels
Classifiers
Experiments

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Magdy, W., Sajjad, H., El-Ganainy, T., & Sebastiani, F. (2015). Distant supervision for tweet classification using youtube labels. In Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015 (pp. 638-641). AAAI Press.

Distant supervision for tweet classification using youtube labels. / Magdy, Walid; Sajjad, Hassan; El-Ganainy, Tarek; Sebastiani, Fabrizio.

Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015. AAAI Press, 2015. p. 638-641.

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

Magdy, W, Sajjad, H, El-Ganainy, T & Sebastiani, F 2015, Distant supervision for tweet classification using youtube labels. in Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015. AAAI Press, pp. 638-641, 9th International Conference on Web and Social Media, ICWSM 2015, Oxford, United Kingdom, 26/5/15.
Magdy W, Sajjad H, El-Ganainy T, Sebastiani F. Distant supervision for tweet classification using youtube labels. In Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015. AAAI Press. 2015. p. 638-641
Magdy, Walid ; Sajjad, Hassan ; El-Ganainy, Tarek ; Sebastiani, Fabrizio. / Distant supervision for tweet classification using youtube labels. Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015. AAAI Press, 2015. pp. 638-641
@inproceedings{d2acd56b8dde4fb1a4521fe4d8cbbe4b,
title = "Distant supervision for tweet classification using youtube labels",
abstract = "We study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another. In particular, we apply classes assigned to YouTube videos to tweet slinking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The experiments we have runs how that a tweet classifier trained via these automatically labelled data substantially outperforms an analogous classifier trained with a limited amount of manually labelled data.",
author = "Walid Magdy and Hassan Sajjad and Tarek El-Ganainy and Fabrizio Sebastiani",
year = "2015",
language = "English",
isbn = "9781577357339",
pages = "638--641",
booktitle = "Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015",
publisher = "AAAI Press",

}

TY - GEN

T1 - Distant supervision for tweet classification using youtube labels

AU - Magdy, Walid

AU - Sajjad, Hassan

AU - El-Ganainy, Tarek

AU - Sebastiani, Fabrizio

PY - 2015

Y1 - 2015

N2 - We study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another. In particular, we apply classes assigned to YouTube videos to tweet slinking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The experiments we have runs how that a tweet classifier trained via these automatically labelled data substantially outperforms an analogous classifier trained with a limited amount of manually labelled data.

AB - We study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another. In particular, we apply classes assigned to YouTube videos to tweet slinking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The experiments we have runs how that a tweet classifier trained via these automatically labelled data substantially outperforms an analogous classifier trained with a limited amount of manually labelled data.

UR - http://www.scopus.com/inward/record.url?scp=84960970232&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960970232&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84960970232

SN - 9781577357339

SP - 638

EP - 641

BT - Proceedings of the 9th International Conference on Web and Social Media, ICWSM 2015

PB - AAAI Press

ER -