fBERT: A Neural Transformer for Identifying Offensive Content

Diptanu Sarkar; Marcos Zampieri; Tharindu Ranasinghe; Alexander Ororbia

doi:10.18653/v1/2021.findings-emnlp.154

fBERT: A Neural Transformer for Identifying Offensive Content

Diptanu Sarkar, Marcos Zampieri, Tharindu Ranasinghe, Alexander Ororbia

Computer Science

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

Transformer-based models such as BERT, XLNET, and XLM-R have achieved state-of-the-art performance across various NLP tasks including the identification of offensive language and hate speech, an important problem in social media. In this paper, we present fBERT, a BERT model retrained on SOLID, the largest English offensive language identification corpus available with over 1.4 million offensive instances. We evaluate fBERT’s performance on identifying offensive content on multiple English datasets and we test several thresholds for selecting instances from SOLID. The fBERT model will be made freely available to the community.

Original language	English
Title of host publication	Findings of the Association for Computational Linguistics: EMNLP 2021
Publisher	Association for Computational Linguistics (ACL)
Pages	1792-1798
DOIs	https://doi.org/10.18653/v1/2021.findings-emnlp.154
Publication status	Published - Nov 2021
Event	The 2021 Conference on Empirical Methods in Natural Language Processing - Punta Cana, Dominican Republic Duration: 7 Nov 2021 → 11 Nov 2021 https://2021.emnlp.org/

Conference

Conference	The 2021 Conference on Empirical Methods in Natural Language Processing
Abbreviated title	EMNLP 2021
Country/Territory	Dominican Republic
City	Punta Cana
Period	7/11/21 → 11/11/21
Internet address	https://2021.emnlp.org/

Access to Document

10.18653/v1/2021.findings-emnlp.154

Cite this

@inproceedings{59c8cf62bce94a66b4cef97d2456e185,

title = "fBERT: A Neural Transformer for Identifying Offensive Content",

abstract = "Transformer-based models such as BERT, XLNET, and XLM-R have achieved state-of-the-art performance across various NLP tasks including the identification of offensive language and hate speech, an important problem in social media. In this paper, we present fBERT, a BERT model retrained on SOLID, the largest English offensive language identification corpus available with over 1.4 million offensive instances. We evaluate fBERT{\textquoteright}s performance on identifying offensive content on multiple English datasets and we test several thresholds for selecting instances from SOLID. The fBERT model will be made freely available to the community.",

author = "Diptanu Sarkar and Marcos Zampieri and Tharindu Ranasinghe and Alexander Ororbia",

year = "2021",

month = nov,

doi = "10.18653/v1/2021.findings-emnlp.154",

language = "English",

pages = "1792--1798",

booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",

publisher = "Association for Computational Linguistics (ACL)",

address = "United States",

note = "The 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021 ; Conference date: 07-11-2021 Through 11-11-2021",

url = "https://2021.emnlp.org/",

}

Sarkar, D, Zampieri, M, Ranasinghe, T & Ororbia, A 2021, fBERT: A Neural Transformer for Identifying Offensive Content. in Findings of the Association for Computational Linguistics: EMNLP 2021. Association for Computational Linguistics (ACL), pp. 1792-1798, The 2021 Conference on Empirical Methods in Natural Language Processing, Punta Cana, Dominican Republic, 7/11/21. https://doi.org/10.18653/v1/2021.findings-emnlp.154

TY - GEN

T1 - fBERT: A Neural Transformer for Identifying Offensive Content

AU - Sarkar, Diptanu

AU - Zampieri, Marcos

AU - Ranasinghe, Tharindu

AU - Ororbia, Alexander

PY - 2021/11

Y1 - 2021/11

N2 - Transformer-based models such as BERT, XLNET, and XLM-R have achieved state-of-the-art performance across various NLP tasks including the identification of offensive language and hate speech, an important problem in social media. In this paper, we present fBERT, a BERT model retrained on SOLID, the largest English offensive language identification corpus available with over 1.4 million offensive instances. We evaluate fBERT’s performance on identifying offensive content on multiple English datasets and we test several thresholds for selecting instances from SOLID. The fBERT model will be made freely available to the community.

AB - Transformer-based models such as BERT, XLNET, and XLM-R have achieved state-of-the-art performance across various NLP tasks including the identification of offensive language and hate speech, an important problem in social media. In this paper, we present fBERT, a BERT model retrained on SOLID, the largest English offensive language identification corpus available with over 1.4 million offensive instances. We evaluate fBERT’s performance on identifying offensive content on multiple English datasets and we test several thresholds for selecting instances from SOLID. The fBERT model will be made freely available to the community.

UR - https://aclanthology.org/2021.findings-emnlp.154

U2 - 10.18653/v1/2021.findings-emnlp.154

DO - 10.18653/v1/2021.findings-emnlp.154

M3 - Conference publication

SP - 1792

EP - 1798

BT - Findings of the Association for Computational Linguistics: EMNLP 2021

PB - Association for Computational Linguistics (ACL)

T2 - The 2021 Conference on Empirical Methods in Natural Language Processing

Y2 - 7 November 2021 through 11 November 2021

ER -

fBERT: A Neural Transformer for Identifying Offensive Content

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this