In this digital world, the emergence of social media platforms is slowly increasing day by day. Users who work online now share their information with ease using computers, cell phones etc. However, this has led to an increase in criminal acts like online harassment, bullying, hate speech, and Cyberbullying, especially among youngsters. As a result, it has become a global problem.
The use of electronic means to harass someone by sending harmful messages through social media, instant messaging or digital messages. It turns out to be a group of insults, humiliations that can affect them physically or emotionally and sometimes lead to suicide attempts in a very serious situation. To find out if someone is abusing or not and stop being abused we create a program using a web page and machine learning (Naïve Bayes) idea to find and stop the person sending messages through those words.
To show how the recovery process took place we used a web page in which two people chat and if one person uses bad words the words will be changed to "*" and point to another user and receive a warning to stop using and after that a user limit will be blocked. If a user wants to send information via a URL link then we provide what kind of abuse data to that. All of this is done with a Naïve Bayes algorithm code written in python, a web page with java as a backend and SQL database for keeping conversations. Therefore, with the help of this project, we can identify abusive words and stop bullying and reduce criminal acts.