Scientists have developed a new algorithm that can spot fake accounts on social networking sites such as Facebook and Twitter.
The method is based on the assumption that fake accounts tend to establish improbable links to other users in the networks, according to the study in published in the journal Social Network Analysis and Mining.
"With recent disturbing news about failures to safeguard user privacy, and targeted use of social media by Russia to influence elections, rooting out fake users has never been of greater importance," said Dima Kagan, lead researcher from Ben-Gurion University in Israel.
"We tested our algorithm on simulated and real-world data sets on 10 different social networks and it performed well on both," Kagan said.
The algorithm consists of two main iterations based on machine-learning algorithms. The first constructs a link prediction classifier that can estimate, with high accuracy, the probability of a link existing between two users. The second iteration generates a new set of meta-features based on the features created by the link prediction classifier.
The researchers, including those from University of Washington in the US, used these meta-features and constructed a generic classifier that can detect fake profiles in a variety of online social networks.
"Overall, the results demonstrated that in a real-life friendship scenario we can detect people who have the strongest friendship ties as well as malicious users, even on Twitter," researchers said.
"Our method outperforms other anomaly detection methods and we believe that it has considerable potential for a wide range of applications particularly in the cyber-security arena," they said.
(This story has not been edited by Business Standard staff and is auto-generated from a syndicated feed.)