ChatGuard: A Profanity Classification Approach for Safer Online Conversations

With every passing day, our world becomes more and more technology driven. Kids are now being introduced to a variety of devices and exposed to the internet from a very young age. The internet is full of information, good or bad. And it has become fundamental for parents to exercise some control and censorship over not only the content that their kids consume but also their interactions over the internet. The aim of this project has been to build to tool that will help in doing just that. It allows to closely monitor the language so that any use of obscene or offensive words can be detected. Even the simple knowledge of what words has been used, can aid parents in initiating a conversation to help their kids understand why certain language is inappropriate or even wrong to guide them as they unlearn any such unhealthy habits and language patterns. The program essentially analyses the text we provide to it and detect profanity or offensive terms. It has been trained against a large dataset and uses machine learning methods to process text and find out if there are any such terms in it.