Are You Human? Detecting Bots on Twitter Using BERT

Dissemination of fake news on Twitter is a rapidly growing problem, mostly due to the increasing number of bots. Hence, automatic bot detection is becoming an important area of research. In this work, we present the BERT-based bot detection model along with exploratory data analysis of tweets writte...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA) pp. 631 - 636
Main Authors Dukic, David, Keca, Dominik, Stipic, Dominik
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2020
Subjects
BERT model
Bit error rate
Bot (Internet)
bot detection
Deep learning
emoji2vec
Feature extraction
gender prediction
latent Dirichlet allocation
Logistics
shallow vs. deep learning
Social networking (online)
t-SNE
Task analysis
Online AccessGet full text
DOI10.1109/DSAA49011.2020.00089

Cover

More Information
Summary:Dissemination of fake news on Twitter is a rapidly growing problem, mostly due to the increasing number of bots. Hence, automatic bot detection is becoming an important area of research. In this work, we present the BERT-based bot detection model along with exploratory data analysis of tweets written by bots and humans. We statistically prove that including additional features alongside contextualized embeddings boosts model performance. Furthermore, we develop a gender prediction model using derived features and compare the difficulties of the two tasks. Finally, we demonstrate how Logistic Regression outperforms Deep Neural Network on both tasks.
DOI:10.1109/DSAA49011.2020.00089