Sneha Bushetty,Prasanna Thummalacheruvu,Vineetha Ramavath, C.Jagadeswari,Meghana Devarapalli,





Online social forums are a great place to express one’s opinions on others' work. But due to the threat of harassment and abuse online, many people stop expressing themselves and give up on seeking different opinions. This leads to the complete shutdown of the user comments section in many communities. Hence, there is a need to identify an efficient way to detect the level of toxicity in the comments posted online, which will be helpful to the content moderators who monitor the data obtained from the comments section on online forums. In this paper, we train various machine learning and deep learning models like NB-SVM, LSTM, BERT on the toxic comments dataset and analyze which approach is efficient for the task of classification of toxic comments.


I. Amir Moradibaad1, RaminJalilian Mashhoud2, Use Dimensionality Reduction and SVM Methods to Increase the Penetration Rate of Computer Network, J. Mech. Cont.& Math. Sci., Vol.-14, No.-4, July-August (2019), pp 8-26.

II. Arif Ullah, Umeriqbal, Ijaz Ali Shoukat,Abdul Rauf, O Y Usman,Sheeraz Ahmed, Zeeshan Najam, An Energy-Efficient Task Scheduling using BAT Algorithm for Cloud Computing, Vol.-14, No.-4, J. Mech. Cont.& Math. Sci.,July-August (2019), pp 613-627.
III. Classification of Abusive Comments in Social Media using Deep Learning, Published in 2019 Proceedings of the Third International Conference on Computing Methodologies and Communication (ICCMC 2019) IEEE Xplore
IV. Is preprocessing of text worth your time for toxic comment classification, Int’l Conf. Artificial Intelligence | ICAI’18 |
V. Julian Risch and Ralf KrestelHasso: Toxic Comment Detection in Online Discussions, Plattner Institute, University of Potsdam.
VI. Jacob Devlin,Ming-Wei Chang,Kenton Lee,Kristina Toutanova, Google AI Language: BERT: Pre-training of Deep Bidirectional Transformers forLanguage Understanding
VII. Sidawang and Christofer D.manning: Baselines and Bigrams: Simple, Good Sentiment and Topic Classification, Department of Computer Science Stanford University

View Download