The focus of this project is to create a smart system that can find instances of cyberbullying on social media using sentiment analysis and machine learning approaches. With more online platforms emerging every day, cyberbullying and other negative behaviours are on the rise, negatively affecting the safety of users and their mental health. Using Natural Language Processing (NLP), the proposed system will use pre-processed text data to extract relevant feature sets via TF-IDF, and will have the ability to use sentiment analysis to assess the feelings of the users creating the content in order to enhance the model's ability to differentiate between bullicious and non-bullicious content. In addition, the project utilizes multiple types of machine learning algorithms, such as Naive Bayes, Logistic Regression, Support Vector Machine and Random Forest, having them evaluated based on their performance in categorising content associated with cyberbullying in near real time. By automating the identification of cyberbullying and deterring individuals from engaging in these harmful forms of behaviour on social media, this project aims to create a safer place for everyone to enjoy all that social media has to offer.
sivamani et al. (Thu,) studied this question.