Hundreds of millions of people use some of the countless social networking sites while billions use those and the bigger, more well-known, sites. A research team based in India and Saudi Arabia reports a new approach to detecting fake accounts on social media sites in the International Journal of Internet Technology and Secured Transactions.
Srinivas Rao of the Department of CSE at JNTUK in Kakinada, Gugulothu Narsimha of the Department of CSE, at JNTUH in Hyderabad, India, and Jayadev Gyani of Majmaah University in Saudi Arabia, explain that there are millions of fake accounts on social media sites. Some of them may well be entirely innocuous, while others are run by scammers, spammers, and those intent on spreading disinformation whether medical, scientific, political, or indeed in any other realm of human endeavour.
“Fake accounts are created for profitable malicious activities, such as spamming, click-fraud, malware distribution, and identity fraud,” the team explains. “Some fakes are created to increase the visibility of niche content, forum posts, and fan pages by manipulating votes/view counts. People also create fake profiles for social reasons and it includes the friendly pranks, stalking, cyberbullying, and concealing a real identity to bypass real-life constraints,” they add.
In their new work, the team describes an optimal validation model that uses a multi-swarm fruit fly algorithm to home in on the fake accounts once trained. This fuzzy logic approach can readily differentiate between genuine and fake accounts with a view to improving the overall trustworthiness of online identities. The team has demonstrated in their proof of principle efficacy when faced with fake accounts on the Facebook and Google+ social networks.
Rao, P.S., Gyani, J. and Narsimha, G. (2021) ‘OVM-OSN: an optimal validation model applied to detection of fake accounts on online social networks’, Int. J. Internet Technology and Secured Transactions, Vol. 11, No. 2, pp.109–130.