Social Media Analysis to Identify Friend or Foe

Abstract

The Data generated online is increasing at a breakneck pace. With over 315 million monthly active users and over 502 million tweets per day, Social Media platforms have become a source of significant data generation that, if analyzed, can be used to prevent cyber-attacks. Social media platforms are used by enemy agencies and banned organi zations for attacking defense personnel. Identifying these personnels on social media will help safeguard assets and avoid information lapses. Using manual searches, this identification of profiles is laborious, manpower-consuming, and inefficient. Identifying friend or foe on social media can be complex due to the inherent nature of online interactions, where individuals may hide their true identities and motivations. However, some certain strategies and indicators can help users distinguish between gen uine friends and potential adversaries. One area where social media, particularly Twit ter, has made a significant impact is sentiment analysis. Sentiment analysis involves the process of gauging the emotions, opinions, and attitudes expressed in text, in this case, tweets. The aim of this research is to automate the process of identifying friend or foe from the tweets extracted using Twitter.Identifying Friend or Foe is based on identifying excessive positive or excessive negativity and aggression in online interactions. In this research, it is shown that the users can be clustered into friend or foe based on the contents of the tweets and the same can be used for any trending topic in general. These users can then be monitored for their activities on social media, which can help prevent social media crimes.