From Tweets to Tolerance: Empowering Cyberbullying Detection with Deep Learning Models

Cyberbullying is the spreading of hate among different religions and people through the Internet. With the worldwide growth in Twitter users, it is now more prone to bullying. There are groups, organizations, and individuals who are involved in online bullying. Such people and individuals can be a severe threat to our society. It is the need of the hour to detect bullying and then classify its types to check the severity of such bullying comments. Work has been done to detect online abuse on social media platforms. Machine learning models were used earlier, which proved to be successful in detecting bullying, but they were not very accurate in the case of large datasets. Later, deep learning models using the techniques of natural language processing were introduced, which can detect bullying content accurately. Previous work on text classification has mostly handled binary class problems, and minor work has been done for multi-class classification of cyberbullying. Our study aims to identify online abuse and categorize cyberbullying using the 'Malignant Comment Classification' dataset from Kaggle. We have trained our machine learning and then deep learning models and adopted a supervised learning approach. For evaluating performance on traditional models Support vector machines, K-nearest neighbors, decision trees, and boosting algorithms were applied to the dataset. We used LSTM, bidirectional encoders, and BERT as deep learning techniques. After the training phase, we made a comparison between these two different approaches, i.e., machine learning and deep learning. The experimental results show that DL models are more effective in text classification problems, and the BERT model obtained an accuracy better than other state-of-the-art models.