There has been a significant impact on people's behavior due to misinformation, especially during the Coronavirus Disease 2019, which directly affects general health awareness. Many studies showed that misinformation campaigns could be more widespread and damaging because of the advent of social media. Understanding the danger of fake news and misinformation, many algorithms and methods have been proposed to tackle the problem of fake news detection. The aim of this study is to detect fake news on social media, in particular, on Twitter, by using Transformers. Specifically, instead of classifying tweets that contain fake news or not, the BERT model is used to identify tweets into three categories: tweets that contain fake news, neutral tweets, and tweets that debunk fake news.

There has been a significant impact on people's behavior due to misinformation, especially during the Coronavirus Disease 2019, which directly affects general health awareness. Many studies showed that misinformation campaigns could be more widespread and damaging because of the advent of social media. Understanding the danger of fake news and misinformation, many algorithms and methods have been proposed to tackle the problem of fake news detection. The aim of this study is to detect fake news on social media, in particular, on Twitter, by using Transformers. Specifically, instead of classifying tweets that contain fake news or not, the BERT model is used to identify tweets into three categories: tweets that contain fake news, neutral tweets, and tweets that debunk fake news.

The Influence Of Misinformation On Twitter During COVID-19

LE, NGOC DIEM
2022/2023

Abstract

There has been a significant impact on people's behavior due to misinformation, especially during the Coronavirus Disease 2019, which directly affects general health awareness. Many studies showed that misinformation campaigns could be more widespread and damaging because of the advent of social media. Understanding the danger of fake news and misinformation, many algorithms and methods have been proposed to tackle the problem of fake news detection. The aim of this study is to detect fake news on social media, in particular, on Twitter, by using Transformers. Specifically, instead of classifying tweets that contain fake news or not, the BERT model is used to identify tweets into three categories: tweets that contain fake news, neutral tweets, and tweets that debunk fake news.
2022
The Influence Of Misinformation On Twitter During COVID-19
There has been a significant impact on people's behavior due to misinformation, especially during the Coronavirus Disease 2019, which directly affects general health awareness. Many studies showed that misinformation campaigns could be more widespread and damaging because of the advent of social media. Understanding the danger of fake news and misinformation, many algorithms and methods have been proposed to tackle the problem of fake news detection. The aim of this study is to detect fake news on social media, in particular, on Twitter, by using Transformers. Specifically, instead of classifying tweets that contain fake news or not, the BERT model is used to identify tweets into three categories: tweets that contain fake news, neutral tweets, and tweets that debunk fake news.
Misinformation
Fake News Detection
Twitter
COVID-19
File in questo prodotto:
File Dimensione Formato  
DiemLe_Thesis_final (1).pdf

accesso aperto

Dimensione 10.78 MB
Formato Adobe PDF
10.78 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/50208