A Strategic Review on Offensive Data Identification Over Social Networking Environment Using Machine Learning Techniques

The social networking environment has evolved globally and dramatically in recent years. Peoples in the world needs a social networking platform to quickly communicate and share their thoughts, views, and ideas with others all over the world, but there are many defects that are happening because of using these communities. A novel learning scheme is required to resolve the issue of posting offensive data on social media. In the previous research, many of the standard machine learning models were enforced for the automatic determination of bullying on social media. However, the models have solved some of the mandatory features that can be used to detect or classify a comment as online harassment that negatively impacts others. This is also known as “online bullying.” The social network environment is a best online tool for everyone, but the devastating effects of such unpleasant information posting issues bring the platform down. This study attempts to provide an overview of various tactics employed in past research, with a focus on identifying or detecting online bullying through the utilization of information from online social networking sites. Finally, a detailed discussion about the open issues for future research in the area of social networking sites based on identifying or classifying offensive content in the user's post using advanced machine learning techniques was organized to provide complete stability from false negative situations for both users and the social networking environment.