Twitter Data Analysis to Discover Suspicious Users and Malicious Content
Praveen Rao; Charles Kamhoua; Laurent Njilla; Kevin Kwiat
Summary
The power of social media is undeniable: may it be in a marketing or political campaign, sharing breaking news, or during catastrophic events. Unfortunately, social media has also become a major weapon for launching cyber attacks on an organization and its people. By hacking into accounts of (popular) users, hackers can post false information, which can go viral and lead to economic damages and create havoc among people. Another major threat on social media is the spread of malware through social media posts by tricking innocent users to click unsuspecting links. Due to these reasons, organizations are developing policies for usage of social media and investing a lot of money and resources to secure their infrastructure and prevent such attacks.
The invention comprises a system and article of manufacture to discover potential cyber (and other) threats on Twitter. The invention provides a unified framework for modeling and reasoning about the veracity of tweets to discover suspicious users and malicious content. The invention builds on the concept of Markov logic networks (MLNs) for knowledge representation and reasoning under uncertainty.
Markets
Cyber Security, Data Analytics, Information Security
IP Status
This technology is the subject of a US patent granted on July 9, 2019. Rights are assigned to the United States Air Force but should be available for licensing.
Key Words
Twitter, Fake News, Cyber Security, Data Analytics
Licensing
Technology is available for licensing through the US Air Force. Contact the Caesar Group to help connect you to the Tech Transfer agent at the responsible Air Force organization.
References