@inproceedings{harrag-etal-2020-bert,
title = "Bert Transformer model for Detecting {A}rabic {GPT}2 Auto-Generated Tweets",
author = "Harrag, Fouzi and
Dabbah, Maria and
Darwish, Kareem and
Abdelali, Ahmed",
editor = "Zitouni, Imed and
Abdul-Mageed, Muhammad and
Bouamor, Houda and
Bougares, Fethi and
El-Haj, Mahmoud and
Tomeh, Nadi and
Zaghouani, Wajdi",
booktitle = "Proceedings of the Fifth Arabic Natural Language Processing Workshop",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.wanlp-1.19",
pages = "207--214",
abstract = "During the last two decades, we have progressively turned to the Internet and social media to find news, entertain conversations and share opinion. Recently, OpenAI has developed a machine learning system called GPT-2 for Generative Pre-trained Transformer-2, which can produce deepfake texts. It can generate blocks of text based on brief writing prompts that look like they were written by humans, facilitating the spread false or auto-generated text. In line with this progress, and in order to counteract potential dangers, several methods have been proposed for detecting text written by these language models. In this paper, we propose a transfer learning based model that will be able to detect if an Arabic sentence is written by humans or automatically generated by bots. Our dataset is based on tweets from a previous work, which we have crawled and extended using the Twitter API. We used GPT2-Small-Arabic to generate fake Arabic Sentences. For evaluation, we compared different recurrent neural network (RNN) word embeddings based baseline models, namely: LSTM, BI-LSTM, GRU and BI-GRU, with a transformer-based model. Our new transfer-learning model has obtained an accuracy up to 98{\%}. To the best of our knowledge, this work is the first study where ARABERT and GPT2 were combined to detect and classify the Arabic auto-generated texts.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="harrag-etal-2020-bert">
<titleInfo>
<title>Bert Transformer model for Detecting Arabic GPT2 Auto-Generated Tweets</title>
</titleInfo>
<name type="personal">
<namePart type="given">Fouzi</namePart>
<namePart type="family">Harrag</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Dabbah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kareem</namePart>
<namePart type="family">Darwish</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="family">Abdelali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifth Arabic Natural Language Processing Workshop</title>
</titleInfo>
<name type="personal">
<namePart type="given">Imed</namePart>
<namePart type="family">Zitouni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="family">Abdul-Mageed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fethi</namePart>
<namePart type="family">Bougares</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mahmoud</namePart>
<namePart type="family">El-Haj</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nadi</namePart>
<namePart type="family">Tomeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wajdi</namePart>
<namePart type="family">Zaghouani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Barcelona, Spain (Online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>During the last two decades, we have progressively turned to the Internet and social media to find news, entertain conversations and share opinion. Recently, OpenAI has developed a machine learning system called GPT-2 for Generative Pre-trained Transformer-2, which can produce deepfake texts. It can generate blocks of text based on brief writing prompts that look like they were written by humans, facilitating the spread false or auto-generated text. In line with this progress, and in order to counteract potential dangers, several methods have been proposed for detecting text written by these language models. In this paper, we propose a transfer learning based model that will be able to detect if an Arabic sentence is written by humans or automatically generated by bots. Our dataset is based on tweets from a previous work, which we have crawled and extended using the Twitter API. We used GPT2-Small-Arabic to generate fake Arabic Sentences. For evaluation, we compared different recurrent neural network (RNN) word embeddings based baseline models, namely: LSTM, BI-LSTM, GRU and BI-GRU, with a transformer-based model. Our new transfer-learning model has obtained an accuracy up to 98%. To the best of our knowledge, this work is the first study where ARABERT and GPT2 were combined to detect and classify the Arabic auto-generated texts.</abstract>
<identifier type="citekey">harrag-etal-2020-bert</identifier>
<location>
<url>https://aclanthology.org/2020.wanlp-1.19</url>
</location>
<part>
<date>2020-12</date>
<extent unit="page">
<start>207</start>
<end>214</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Bert Transformer model for Detecting Arabic GPT2 Auto-Generated Tweets
%A Harrag, Fouzi
%A Dabbah, Maria
%A Darwish, Kareem
%A Abdelali, Ahmed
%Y Zitouni, Imed
%Y Abdul-Mageed, Muhammad
%Y Bouamor, Houda
%Y Bougares, Fethi
%Y El-Haj, Mahmoud
%Y Tomeh, Nadi
%Y Zaghouani, Wajdi
%S Proceedings of the Fifth Arabic Natural Language Processing Workshop
%D 2020
%8 December
%I Association for Computational Linguistics
%C Barcelona, Spain (Online)
%F harrag-etal-2020-bert
%X During the last two decades, we have progressively turned to the Internet and social media to find news, entertain conversations and share opinion. Recently, OpenAI has developed a machine learning system called GPT-2 for Generative Pre-trained Transformer-2, which can produce deepfake texts. It can generate blocks of text based on brief writing prompts that look like they were written by humans, facilitating the spread false or auto-generated text. In line with this progress, and in order to counteract potential dangers, several methods have been proposed for detecting text written by these language models. In this paper, we propose a transfer learning based model that will be able to detect if an Arabic sentence is written by humans or automatically generated by bots. Our dataset is based on tweets from a previous work, which we have crawled and extended using the Twitter API. We used GPT2-Small-Arabic to generate fake Arabic Sentences. For evaluation, we compared different recurrent neural network (RNN) word embeddings based baseline models, namely: LSTM, BI-LSTM, GRU and BI-GRU, with a transformer-based model. Our new transfer-learning model has obtained an accuracy up to 98%. To the best of our knowledge, this work is the first study where ARABERT and GPT2 were combined to detect and classify the Arabic auto-generated texts.
%U https://aclanthology.org/2020.wanlp-1.19
%P 207-214
Markdown (Informal)
[Bert Transformer model for Detecting Arabic GPT2 Auto-Generated Tweets](https://aclanthology.org/2020.wanlp-1.19) (Harrag et al., WANLP 2020)
ACL