NLP and Machine Learning Techniques for Detecting Insulting Comments on Social Networking Platforms

Abstract

In the era of social media and networking, the usage of bad words and aggressive words has been increased significantly. The young population is playing a major role in it. Cyberbullying affects more than half of the young population using social media. Insults in social media websites create negative interactions within the network. These remarks build up a culture of disrespect in cyberspace. Algorithms and tools used to understand and mitigate it are mostly inactive. Also, current implementations on insult detection using machine learning and natural language processing have very low recall rates. In short, the paper involves determining ways to identify bullying in text by analyzing and experimenting with different methods to find the feasible way of classifying such comments. We proposed a efficient algorithm to identify the bullying test and aggressive comments and analyses these comments to check the validity. NLP and Machine learning is used for analyzing the social comment and identified the aggressive effect of an individual or a group. An effective classifier acts as the core component in a final prototype system that can detect cyberbullying on social media.