Detecting Suspicious News over Social Media using Ant Colony Optimization

Abstract

The online social media platforms has become the trending as it provide the convenient and free access to users to share their day to day activities and other information. Despite the significance of these online social media platforms, there are also the people that can mislead the other by posting the fake news. These kinds of news are termed as suspicious news. Such kind of misleading news can badly affect the society. It is the way too hard to completely restrict such people from posting anything on social media. But after the detection of such activities, these posts can be removed from social media and users can be restricted from the respective platform. From the years, researchers are continuously working to detect the suspicious activities using machine learning and data mining techniques. This research work addresses the problem of suspicious news detection using the ant colony optimization based ant miner plus technique. This proposed approach is termed as ACODSN (Ant Colony Optimization for the Detection of Suspicious News). The experimentation is conducted on the dataset of FakenewsNet. The system performance is analyzed in terms of evaluation metrics of recall, precision, and f-measure.