Identifying the Authors’ National Variety of English in Social Media Texts

Vasiliki Simaki; Panagiotis Simakis; Carita Paradis; Kerren Andreas

doi:10.26615/978-954-452-049-6_086

Identifying the Authors’ National Variety of English in Social Media Texts

Vasiliki Simaki, Panagiotis Simakis, Carita Paradis, Kerren Andreas

English Studies

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

Abstract

In this paper, we present a study for the identification of authors’ national variety of English in texts from social media. In data from Facebook and Twitter, information about the author’s social profile is annotated, and the national English variety (US, UK, AUS, CAN, NNS) that each author uses is attributed. We tested four feature types: formal linguistic features, POS features, lexicon-based features related to the different varieties, and data-based features from each English variety. We used various machine learning algorithms for the classification experiments, and we implemented a feature selection
process. The classification accuracy achieved, when the 31 highest ranked
features were used, was up to 77.32%. The experimental results are evaluated, and the efficacy of the ranked features discussed.

Original language	English
Title of host publication	Recent Advances in Natural Language Processing
Subtitle of host publication	Proceedings
Editors	Galia Angelova, Kalina Bontcheva, Ruslan Metkov, Ivelina Nikolova, Irina Temnikova
Place of Publication	Varna
Publisher	Association for Computational Linguistics
Pages	671-678
ISBN (Electronic)	978-954-452-049-6
ISBN (Print)	978-954-452-048-9
DOIs	https://doi.org/10.26615/978-954-452-049-6_086
Publication status	Published - 2017
Event	The 11th Biennial Conference on Recent Advances In Natural Language Processing (RANLP '17), 2-8 September 2017, Varna, Bulgaria - Varna, Bulgaria Duration: 2017 Sept 2 → 2017 Sept 8 http://lml.bas.bg/ranlp2017/start.php

Conference

Conference	The 11th Biennial Conference on Recent Advances In Natural Language Processing (RANLP '17), 2-8 September 2017, Varna, Bulgaria
Abbreviated title	RANLP '17
Country/Territory	Bulgaria
City	Varna
Period	2017/09/02 → 2017/09/08
Internet address	http://lml.bas.bg/ranlp2017/start.php

Subject classification (UKÄ)

General Language Studies and Linguistics

Access to Document

10.26615/978-954-452-049-6_086

Cite this

Simaki, V., Simakis, P., Paradis, C., & Andreas, K. (2017). Identifying the Authors’ National Variety of English in Social Media Texts. In G. Angelova, K. Bontcheva, R. Metkov, I. Nikolova, & I. Temnikova (Eds.), Recent Advances in Natural Language Processing: Proceedings (pp. 671-678). Association for Computational Linguistics. https://doi.org/10.26615/978-954-452-049-6_086

Simaki, Vasiliki ; Simakis, Panagiotis ; Paradis, Carita et al. / Identifying the Authors’ National Variety of English in Social Media Texts. Recent Advances in Natural Language Processing: Proceedings. editor / Galia Angelova ; Kalina Bontcheva ; Ruslan Metkov ; Ivelina Nikolova ; Irina Temnikova. Varna : Association for Computational Linguistics, 2017. pp. 671-678

@inproceedings{d6aee74bcf44485eaaaf9c7f9b1947d9,

title = "Identifying the Authors{\textquoteright} National Variety of English in Social Media Texts",

abstract = "In this paper, we present a study for the identification of authors{\textquoteright} national variety of English in texts from social media. In data from Facebook and Twitter, information about the author{\textquoteright}s social profile is annotated, and the national English variety (US, UK, AUS, CAN, NNS) that each author uses is attributed. We tested four feature types: formal linguistic features, POS features, lexicon-based features related to the different varieties, and data-based features from each English variety. We used various machine learning algorithms for the classification experiments, and we implemented a feature selectionprocess. The classification accuracy achieved, when the 31 highest rankedfeatures were used, was up to 77.32%. The experimental results are evaluated, and the efficacy of the ranked features discussed.",

author = "Vasiliki Simaki and Panagiotis Simakis and Carita Paradis and Kerren Andreas",

year = "2017",

doi = "10.26615/978-954-452-049-6_086",

language = "English",

isbn = "978-954-452-048-9",

pages = "671--678",

editor = "Galia Angelova and Kalina Bontcheva and Ruslan Metkov and Ivelina Nikolova and Irina Temnikova",

booktitle = "Recent Advances in Natural Language Processing",

publisher = "Association for Computational Linguistics",

address = "United States",

note = "The 11th Biennial Conference on Recent Advances In Natural Language Processing (RANLP '17), 2-8 September 2017, Varna, Bulgaria , RANLP '17 ; Conference date: 02-09-2017 Through 08-09-2017",

url = "http://lml.bas.bg/ranlp2017/start.php",

}

Simaki, V, Simakis, P, Paradis, C & Andreas, K 2017, Identifying the Authors’ National Variety of English in Social Media Texts. in G Angelova, K Bontcheva, R Metkov, I Nikolova & I Temnikova (eds), Recent Advances in Natural Language Processing: Proceedings. Association for Computational Linguistics, Varna, pp. 671-678, The 11th Biennial Conference on Recent Advances In Natural Language Processing (RANLP '17), 2-8 September 2017, Varna, Bulgaria , Varna, Bulgaria, 2017/09/02. https://doi.org/10.26615/978-954-452-049-6_086

Identifying the Authors’ National Variety of English in Social Media Texts. / Simaki, Vasiliki; Simakis, Panagiotis; Paradis, Carita et al.
Recent Advances in Natural Language Processing: Proceedings. ed. / Galia Angelova; Kalina Bontcheva; Ruslan Metkov; Ivelina Nikolova; Irina Temnikova. Varna: Association for Computational Linguistics, 2017. p. 671-678.

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

TY - GEN

T1 - Identifying the Authors’ National Variety of English in Social Media Texts

AU - Simaki, Vasiliki

AU - Simakis, Panagiotis

AU - Paradis, Carita

AU - Andreas, Kerren

PY - 2017

Y1 - 2017

N2 - In this paper, we present a study for the identification of authors’ national variety of English in texts from social media. In data from Facebook and Twitter, information about the author’s social profile is annotated, and the national English variety (US, UK, AUS, CAN, NNS) that each author uses is attributed. We tested four feature types: formal linguistic features, POS features, lexicon-based features related to the different varieties, and data-based features from each English variety. We used various machine learning algorithms for the classification experiments, and we implemented a feature selectionprocess. The classification accuracy achieved, when the 31 highest rankedfeatures were used, was up to 77.32%. The experimental results are evaluated, and the efficacy of the ranked features discussed.

AB - In this paper, we present a study for the identification of authors’ national variety of English in texts from social media. In data from Facebook and Twitter, information about the author’s social profile is annotated, and the national English variety (US, UK, AUS, CAN, NNS) that each author uses is attributed. We tested four feature types: formal linguistic features, POS features, lexicon-based features related to the different varieties, and data-based features from each English variety. We used various machine learning algorithms for the classification experiments, and we implemented a feature selectionprocess. The classification accuracy achieved, when the 31 highest rankedfeatures were used, was up to 77.32%. The experimental results are evaluated, and the efficacy of the ranked features discussed.

U2 - 10.26615/978-954-452-049-6_086

DO - 10.26615/978-954-452-049-6_086

M3 - Paper in conference proceeding

SN - 978-954-452-048-9

SP - 671

EP - 678

BT - Recent Advances in Natural Language Processing

A2 - Angelova, Galia

A2 - Bontcheva, Kalina

A2 - Metkov, Ruslan

A2 - Nikolova, Ivelina

A2 - Temnikova, Irina

PB - Association for Computational Linguistics

CY - Varna

T2 - The 11th Biennial Conference on Recent Advances In Natural Language Processing (RANLP '17), 2-8 September 2017, Varna, Bulgaria

Y2 - 2 September 2017 through 8 September 2017

ER -

Identifying the Authors’ National Variety of English in Social Media Texts

Abstract

Conference

Subject classification (UKÄ)

Access to Document

Fingerprint

Cite this