Cyberbullying Prediction Using K-Means Clustering with Learning Methodologies for Emotion Detection in Tweets

The rise of social media has led to an increase in real-time data that can reveal potential cyber threats. This research explores the use of emotion detection in tweets to predict cyber attacks, employing K-Means Clustering in combination with machine learning methodologies. By analyzing emotional trends, such as fear, anger, and anxiety, this study establishes a correlation between social media sentiment and cyber attack occurrences. The results demonstrate the potential for early warning systems that harness emotional patterns for cybersecurity. Social media platforms like Twitter provide a wealth of real-time data that can reveal societal trends, including indicators of potential cyber threats. This study investigates the use of emotion detection in tweets as a predictive tool for cyber attack forecasting. By employing Natural Language Processing (NLP) techniques, emotional states such as fear, anger, and anxiety are extracted and analyzed. K-Means Clustering is utilized to group similar emotional patterns, enabling the identification of significant trends. These clusters, combined with historical cyberbullying data, are used as features in supervised learning models for cyber threat prediction. The results demonstrate the effectiveness of integrating unsupervised clustering and machine learning to correlate spikes in negative emotions with the likelihood of cyber incidents. This research highlights the potential for proactive cybersecurity strategies that leverage social media sentiment as an early warning system.