DETECTION OF CYBERBULLYING ON SOCIAL MEDIA USING SVM

COMPUTER SCIENCE ENGINEERING

In order to identify Twitter attacks, this project provides a survey of the twitter stream. Online social networking has grown to be a fairly common means for individuals to connect and communicate. Users spend a lot of time on well- known social networking sites like Facebook, MySpace, or Twitter, where they store and share a plethora of personal data. Our findings indicate that the bulk of clicks come from direct sources, and that spammers use well-known websites to cross- post links to gain additional attention. Apply this method to help with the issue of online spam detection with little overhead even though campaign identification has been utilized for offline spam filtering.We gather a large number of tweets from the public timeline on Twitter and use them to create a statistical classifier. Evaluation findings demonstrate that our classifier swiftly and accurately identifies suspicious URLs. This method looks for correlations between URL redirect chain that were taken froma numbers of tweet and creates commonly shared URLs. because attackers typically waste their limited resources. Cyber bullying, Machine learning, Natural language processing, Social media.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.