A Novel Approach for Social Media Content Filtering Using Machine Learning Technique

A completely interconnected world is possible because of social networks, which allow individuals to interact, share ideas, and organize themselves into digital environments. Both online behavior and online content processing are critical for security applications to be successful. Extremists now have an easy way to spread their ideas and beliefs via social media. Cyberbullying and the spread of false news and phoney reviews on social media are two of the numerous security risks that arise as a result of this. This necessitates the need for the creation of a method to identify and minimize dangerous information on social networks. Risks and the extent of their repercussions are subjectively evaluated in the context of security occurrences. Societal consequences of negative impressions can be dire. Citizens’ polls are routinely used to measure these sentiments, but they take a long time and don’t adjust well to shifting security dynamics. In light of this, we developed a machine learning-based social media content filtering strategy. In order to train our model to identify fraudulent tweets on the Twitter network, we used the four different machine learning approach to find the malicious comments.