
A Hybrid Approach for Detecting Automated Spammers in Twitter
Abstract of Hybrid Approach for Detecting Automated
Hybrid Approach for Detecting Automated
are included in the list of the top 10 websites1 around the worldwide.
Twitter allows the users to follow their favorite politicians, athletes, celebrities, and news channels,
and to subscribe to their content without any hindrance. Through following activity, a follower can receive status updates of subscribed account. Although Twitter and other OSNs are mainly used for various benign purposes, their open nature, huge user base, and real-time message proliferation have made them lucrative targets for cyber criminals and socialbots.
Conclusion
In this paper, we have proposed a hybrid approach exploiting community-based features with metadata-, content-, and interaction-based features for detecting automated spammers in Twitter.
Spammers are generally planted in OSNs for varied purposes,
but absence of real-life identity hinders them to join the trust network of benign users.
Therefore, spammers randomly follow a number of users, but rarely followed back by them, which results in low edge density among their followers and followings.
This type of spammers interaction pattern can be exploited for the development of effective spammers detection systems.