@inproceedings{nafis-etal-2023-towards,
title = "Towards Safer Communities: Detecting Aggression and Offensive Language in Code-Mixed Tweets to Combat Cyberbullying",
author = "Nafis, Nazia and
Kanojia, Diptesh and
Saini, Naveen and
Murthy, Rudra",
editor = {Chung, Yi-ling and
R{{\textbackslash}"ottger}, Paul and
Nozza, Debora and
Talat, Zeerak and
Mostafazadeh Davani, Aida},
booktitle = "The 7th Workshop on Online Abuse and Harms (WOAH)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.woah-1.3",
doi = "10.18653/v1/2023.woah-1.3",
pages = "29--41",
abstract = "Cyberbullying is a serious societal issue widespread on various channels and platforms, particularly social networking sites. Such platforms have proven to be exceptionally fertile grounds for such behavior. The dearth of high-quality training data for multilingual and low-resource scenarios, data that can accurately capture the nuances of social media conversations, often poses a roadblock to this task. This paper attempts to tackle cyberbullying, specifically its two most common manifestations - aggression and offensiveness. We present a novel, manually annotated dataset of a total of 10,000 English and Hindi-English code-mixed tweets, manually annotated for aggression detection and offensive language detection tasks. Our annotations are supported by inter-annotator agreement scores of 0.67 and 0.74 for the two tasks, indicating substantial agreement. We perform comprehensive fine-tuning of pre-trained language models (PTLMs) using this dataset to check its efficacy. Our challenging test sets show that the best models achieve macro F1-scores of 67.87 and 65.45 on the two tasks, respectively. Further, we perform cross-dataset transfer learning to benchmark our dataset against existing aggression and offensive language datasets. We also present a detailed quantitative and qualitative analysis of errors in prediction, and with this paper, we publicly release the novel dataset, code, and models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="nafis-etal-2023-towards">
<titleInfo>
<title>Towards Safer Communities: Detecting Aggression and Offensive Language in Code-Mixed Tweets to Combat Cyberbullying</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nazia</namePart>
<namePart type="family">Nafis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Diptesh</namePart>
<namePart type="family">Kanojia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naveen</namePart>
<namePart type="family">Saini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rudra</namePart>
<namePart type="family">Murthy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>The 7th Workshop on Online Abuse and Harms (WOAH)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yi-ling</namePart>
<namePart type="family">Chung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">R\textbackslash”ottger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debora</namePart>
<namePart type="family">Nozza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zeerak</namePart>
<namePart type="family">Talat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aida</namePart>
<namePart type="family">Mostafazadeh Davani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Cyberbullying is a serious societal issue widespread on various channels and platforms, particularly social networking sites. Such platforms have proven to be exceptionally fertile grounds for such behavior. The dearth of high-quality training data for multilingual and low-resource scenarios, data that can accurately capture the nuances of social media conversations, often poses a roadblock to this task. This paper attempts to tackle cyberbullying, specifically its two most common manifestations - aggression and offensiveness. We present a novel, manually annotated dataset of a total of 10,000 English and Hindi-English code-mixed tweets, manually annotated for aggression detection and offensive language detection tasks. Our annotations are supported by inter-annotator agreement scores of 0.67 and 0.74 for the two tasks, indicating substantial agreement. We perform comprehensive fine-tuning of pre-trained language models (PTLMs) using this dataset to check its efficacy. Our challenging test sets show that the best models achieve macro F1-scores of 67.87 and 65.45 on the two tasks, respectively. Further, we perform cross-dataset transfer learning to benchmark our dataset against existing aggression and offensive language datasets. We also present a detailed quantitative and qualitative analysis of errors in prediction, and with this paper, we publicly release the novel dataset, code, and models.</abstract>
<identifier type="citekey">nafis-etal-2023-towards</identifier>
<identifier type="doi">10.18653/v1/2023.woah-1.3</identifier>
<location>
<url>https://aclanthology.org/2023.woah-1.3</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>29</start>
<end>41</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Towards Safer Communities: Detecting Aggression and Offensive Language in Code-Mixed Tweets to Combat Cyberbullying
%A Nafis, Nazia
%A Kanojia, Diptesh
%A Saini, Naveen
%A Murthy, Rudra
%Y Chung, Yi-ling
%Y R\textbackslash”ottger, Paul
%Y Nozza, Debora
%Y Talat, Zeerak
%Y Mostafazadeh Davani, Aida
%S The 7th Workshop on Online Abuse and Harms (WOAH)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F nafis-etal-2023-towards
%X Cyberbullying is a serious societal issue widespread on various channels and platforms, particularly social networking sites. Such platforms have proven to be exceptionally fertile grounds for such behavior. The dearth of high-quality training data for multilingual and low-resource scenarios, data that can accurately capture the nuances of social media conversations, often poses a roadblock to this task. This paper attempts to tackle cyberbullying, specifically its two most common manifestations - aggression and offensiveness. We present a novel, manually annotated dataset of a total of 10,000 English and Hindi-English code-mixed tweets, manually annotated for aggression detection and offensive language detection tasks. Our annotations are supported by inter-annotator agreement scores of 0.67 and 0.74 for the two tasks, indicating substantial agreement. We perform comprehensive fine-tuning of pre-trained language models (PTLMs) using this dataset to check its efficacy. Our challenging test sets show that the best models achieve macro F1-scores of 67.87 and 65.45 on the two tasks, respectively. Further, we perform cross-dataset transfer learning to benchmark our dataset against existing aggression and offensive language datasets. We also present a detailed quantitative and qualitative analysis of errors in prediction, and with this paper, we publicly release the novel dataset, code, and models.
%R 10.18653/v1/2023.woah-1.3
%U https://aclanthology.org/2023.woah-1.3
%U https://doi.org/10.18653/v1/2023.woah-1.3
%P 29-41
Markdown (Informal)
[Towards Safer Communities: Detecting Aggression and Offensive Language in Code-Mixed Tweets to Combat Cyberbullying](https://aclanthology.org/2023.woah-1.3) (Nafis et al., WOAH 2023)
ACL