Detecting Cyber bullying in Social Media using Deep Learning

This report investigates the challenge of detecting cyberbullying on social media, where traditional methods fall short due to complex, evolving language. A deep learning-based approach is proposed, using NLP techniques and LSTM models to analyse social media posts. Data is preprocessed and classified with a neural network trained on labelled examples of bullying and non-bullying content. The system identifies linguistic patterns such as threats and harassment. Results show improved accuracy and scalability compared to manual methods. The study concludes that intelligent automated detection can enhance content moderation and promote safer online interactions.