Published Paper Details:

In the digital era, providing a safe and respectable online society is a top priority. Cyberbullying, an epidemic that presents a considerable danger to one's mental and social life, is impossible to avoid. This study addresses the issue of cyberbullying classification in Twitter data. We use the simple measures of sentiment analysis to do so in tandem with machine learning. In our project, we employ an SVM to classify tweets as cyberbullying or non-cyberbullying. The sentiment analysis pipeline produced by the sentiment analysis includes text preprocessing, transforming and weighting vectorization, and model training. Moreover, the system employs an interactive widget interface that allows the operators to infiltrate a tweet, and the system processes their input tweets to clean them up using usual expression extraction from URLs and non-alphanumeric characters before passing them to the pre-trained TF-IDF model for training and predicting if it's cyberbullied. Performance evaluation is performed through a classification report which provides precision, recall, F1-score, and support of each class metric. This system has the potential to enable users and administrators to intervene in cyberbullying cases in a preventative manner, making the online community a safer and more inclusive place.