Advanced search
1 file | 106.89 KB Add to list

Towards shared datasets for normalization research

Orphée De Clercq (UGent) , Sarah Schulz (UGent) , Bart Desmet (UGent) and Veronique Hoste (UGent)
Author
Organization
Abstract
In this paper we present a Dutch and English dataset that can serve as a gold standard for evaluating text normalization approaches. With the combination of text messages, message board posts and tweets, these datasets represent a variety of user generated content. All data was manually normalized to their standard form using newly-developed guidelines. We perform automatic lexical normalization experiments on these datasets using statistical machine translation techniques. We focus on both the word and character level and find that we can improve the BLEU score with ca. 20% for both languages. In order for this user generated content data to be released publicly to the research community some issues first need to be resolved. These are discussed in closer detail by focussing on the current legislation and by investigating previous similar data collection projects. With this discussion we hope to shed some light on various difficulties researchers are facing when trying to share social media data.
Keywords
user generated content, resource sharing, text normalization

Downloads

  • 733 Paper.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 106.89 KB

Citation

Please use this url to cite or link to this publication:

MLA
De Clercq, Orphée, et al. “Towards Shared Datasets for Normalization Research.” LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION, edited by Nicoletta Calzolari et al., European Language Resources Association (ELRA), 2014, pp. 1218–23.
APA
De Clercq, O., Schulz, S., Desmet, B., & Hoste, V. (2014). Towards shared datasets for normalization research. In N. Calzolari, K. Choukri, T. Declerck, H. Loftsson, B. Maegaard, J. Mariani, … S. Piperidis (Eds.), LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION (pp. 1218–1223). European Language Resources Association (ELRA).
Chicago author-date
De Clercq, Orphée, Sarah Schulz, Bart Desmet, and Veronique Hoste. 2014. “Towards Shared Datasets for Normalization Research.” In LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION, edited by Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Stelios Piperidis, 1218–23. European Language Resources Association (ELRA).
Chicago author-date (all authors)
De Clercq, Orphée, Sarah Schulz, Bart Desmet, and Veronique Hoste. 2014. “Towards Shared Datasets for Normalization Research.” In LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION, ed by. Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Stelios Piperidis, 1218–1223. European Language Resources Association (ELRA).
Vancouver
1.
De Clercq O, Schulz S, Desmet B, Hoste V. Towards shared datasets for normalization research. In: Calzolari N, Choukri K, Declerck T, Loftsson H, Maegaard B, Mariani J, et al., editors. LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION. European Language Resources Association (ELRA); 2014. p. 1218–23.
IEEE
[1]
O. De Clercq, S. Schulz, B. Desmet, and V. Hoste, “Towards shared datasets for normalization research,” in LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION, Reykjavik, Iceland, 2014, pp. 1218–1223.
@inproceedings{4292521,
  abstract     = {{In this paper we present a Dutch and English dataset that can serve as a gold standard for evaluating text normalization approaches. With the combination of text messages, message board posts and tweets, these datasets represent a variety of user generated content. All data was manually normalized to their standard form using newly-developed guidelines. We perform automatic lexical normalization experiments on these datasets using statistical machine translation techniques. We focus on both the word and character level and find that we can improve the BLEU score with ca. 20% for both languages. In order for this user generated content data to be released publicly to the research community some issues first need to be resolved. These are discussed in closer detail by focussing on the current legislation and by investigating previous similar data collection projects. With this discussion we hope to shed some light on various difficulties researchers are facing when trying to share social media data.}},
  author       = {{De Clercq, Orphée and Schulz, Sarah and Desmet, Bart and Hoste, Veronique}},
  booktitle    = {{LREC 2014 - NINTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION}},
  editor       = {{Calzolari, Nicoletta and Choukri, Khalid and Declerck, Thierry and Loftsson, Hrafn and Maegaard, Bente and Mariani, Joseph and Moreno, Asuncion and Odijk, Jan and Piperidis, Stelios}},
  isbn         = {{9782951740884}},
  keywords     = {{user generated content,resource sharing,text normalization}},
  language     = {{eng}},
  location     = {{Reykjavik, Iceland}},
  pages        = {{1218--1223}},
  publisher    = {{European Language Resources Association (ELRA)}},
  title        = {{Towards shared datasets for normalization research}},
  url          = {{http://www.lrec-conf.org/proceedings/lrec2014/index.html}},
  year         = {{2014}},
}

Web of Science
Times cited: