Research Article
Social Media Toxic-Text Analysis Using Deep Learning Techniques
@INPROCEEDINGS{10.1007/978-3-031-66044-3_21, author={Tripti Agrawal and Shweta Sankhwar and Tanya Chaudhary and Aaditri Saraswat}, title={Social Media Toxic-Text Analysis Using Deep Learning Techniques}, proceedings={Pervasive Knowledge and Collective Intelligence on Web and Social Media. Second EAI International Conference, PerSOM 2023, Hyderabad, India, November 24--25, 2023, Proceedings}, proceedings_a={PERSOM}, year={2024}, month={8}, keywords={Social media Toxic Text analysis Machine Learning Deep-learning Shallow Learning}, doi={10.1007/978-3-031-66044-3_21} }- Tripti Agrawal
Shweta Sankhwar
Tanya Chaudhary
Aaditri Saraswat
Year: 2024
Social Media Toxic-Text Analysis Using Deep Learning Techniques
PERSOM
Springer
DOI: 10.1007/978-3-031-66044-3_21
Abstract
The widespread use of social media in contemporary society has created a vast platform for individuals to express their opinions openly. However, certain anti-social groups misuse this freedom to propagate toxic behavior, including verbal sexual harassment, threats, insults, and obscenities, among others. These behaviors hinder the free exchange of opinions and have led even major social media platforms to limit or disable user comments to counter toxicity. Consequently, the automatic detection and identification of such behavior through machine learning models have become increasingly critical. In this context, this paper examines various machine learning techniques for the classification of toxicity in online comments, utilizing Kaggle’s toxic comment classification dataset. Furthermore, the study assesses the performance of both shallow learning algorithms and deep learning methods, using various evaluation metrics to comprehensively evaluate their effectiveness.
