DEVELOPMENT OF A SENTIMENT ANAYLSIS SYSTEM FOR TWITTER CYBER BULLYING DETECTION

Abstract In today's digital era, social media usage has surged, with individuals frequently sharing experiences on platforms like Twitter. However, this rise in online activity has unfortunately led to an increase in cyber threats, including fraud, blackmail, spam and impersonation. To combat this, a system designed to detect cyber-crimes on Twitter is proposed, aiming to help users identify potential threats and suspicious accounts in real time. This work focuses on enhancing online security by comparing different sentiment analysis techniques for detecting cyberbullying and other online threats. Two distinct machine learning algorithms are used to effectively detect cyberbullying improving awareness of online dangers in both digital and real-world environments.