When Simple n-gram Models Outperform Syntactic Approaches: Discriminating between Dutch and Flemish

Martin Kroon; Masha  Medvedeva; Barbara Plank

When Simple n-gram Models Outperform Syntactic Approaches: Discriminating between Dutch and Flemish

Martin Kroon, Masha Medvedeva, Barbara Plank

Computer Science

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

In this paper we present the results of our participation in the Discriminating between Dutch and Flemish in Subtitles VarDial 2018 shared task. We try techniques proven to work well for discriminating between language varieties as well as explore the potential of using syntactic features, i.e. hierarchical syntactic subtrees. We experiment with different combinations of features. Discriminating between these two languages turned out to be a very hard task, not only for a machine: human performance is only around 0.51 F1 score; our best system is still a simple Naive Bayes model with word unigrams and bigrams. The system achieved an F1 score (macro)
of 0.62, which ranked us 4th in the shared task.

Original language	English
Title of host publication	Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial)
Publisher	Association for Computational Linguistics
Publication date	2018
Pages	244-253
ISBN (Print)	978-1-948087-55-1
Publication status	Published - 2018

Keywords

Dutch-Flemish Discrimination
VarDial 2018
Syntactic Features
Naive Bayes Classifier
Language Variety Classification

Access to Document

DFS_2018Accepted author manuscript, 216 KBLicence: CC BY

Cite this

@inproceedings{a1881a58c2f149aeb21c179f0854feab,

title = "When Simple n-gram Models Outperform Syntactic Approaches: Discriminating between Dutch and Flemish",

abstract = "In this paper we present the results of our participation in the Discriminating between Dutch and Flemish in Subtitles VarDial 2018 shared task. We try techniques proven to work well for discriminating between language varieties as well as explore the potential of using syntactic features, i.e. hierarchical syntactic subtrees. We experiment with different combinations of features. Discriminating between these two languages turned out to be a very hard task, not only for a machine: human performance is only around 0.51 F1 score; our best system is still a simple Naive Bayes model with word unigrams and bigrams. The system achieved an F1 score (macro)of 0.62, which ranked us 4th in the shared task.",

keywords = "Dutch-Flemish Discrimination, VarDial 2018, Syntactic Features, Naive Bayes Classifier, Language Variety Classification",

author = "Martin Kroon and Masha Medvedeva and Barbara Plank",

year = "2018",

language = "English",

isbn = "978-1-948087-55-1",

pages = "244--253",

booktitle = "Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial)",

publisher = "Association for Computational Linguistics",

address = "United States",

}

When Simple n-gram Models Outperform Syntactic Approaches: Discriminating between Dutch and Flemish. / Kroon, Martin; Medvedeva, Masha ; Plank, Barbara.
Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial). Association for Computational Linguistics, 2018. p. 244-253.

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - When Simple n-gram Models Outperform Syntactic Approaches: Discriminating between Dutch and Flemish

AU - Kroon, Martin

AU - Medvedeva, Masha

AU - Plank, Barbara

PY - 2018

Y1 - 2018

N2 - In this paper we present the results of our participation in the Discriminating between Dutch and Flemish in Subtitles VarDial 2018 shared task. We try techniques proven to work well for discriminating between language varieties as well as explore the potential of using syntactic features, i.e. hierarchical syntactic subtrees. We experiment with different combinations of features. Discriminating between these two languages turned out to be a very hard task, not only for a machine: human performance is only around 0.51 F1 score; our best system is still a simple Naive Bayes model with word unigrams and bigrams. The system achieved an F1 score (macro)of 0.62, which ranked us 4th in the shared task.

AB - In this paper we present the results of our participation in the Discriminating between Dutch and Flemish in Subtitles VarDial 2018 shared task. We try techniques proven to work well for discriminating between language varieties as well as explore the potential of using syntactic features, i.e. hierarchical syntactic subtrees. We experiment with different combinations of features. Discriminating between these two languages turned out to be a very hard task, not only for a machine: human performance is only around 0.51 F1 score; our best system is still a simple Naive Bayes model with word unigrams and bigrams. The system achieved an F1 score (macro)of 0.62, which ranked us 4th in the shared task.

KW - Dutch-Flemish Discrimination

KW - VarDial 2018

KW - Syntactic Features

KW - Naive Bayes Classifier

KW - Language Variety Classification

UR - https://aclanthology.coli.uni-saarland.de/papers/W18-3928/w18-3928

UR - http://aclweb.org/anthology/W18-39

M3 - Article in proceedings

SN - 978-1-948087-55-1

SP - 244

EP - 253

BT - Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial)

PB - Association for Computational Linguistics

ER -

When Simple n-gram Models Outperform Syntactic Approaches: Discriminating between Dutch and Flemish

Abstract

Keywords

Access to Document

Fingerprint

Cite this