Defense Mechanism Against Attacks Promoting Spread Of Wrong Information

Abstract

The rise in dependency upon information present on web and social networks has increased the importance of content credibility systems. These sys- tems can help the users to make a right decision related to buying a product, utiliz- ing a service and etc. Application areas including social network blogs of different subjects such as health, food, education, politics and product review/ratings some- times suffer incredible and wrong information flow. The content credibility sys- tems can help users to identify the credible information on these various forums. Recently, researchers have proposed certain mechanisms for these systems, out of which reputation based systems have gained most attention. However, reputa- tion systems are vulnerable to reputation based attacks, like Sybil, Slandering, and Whitewash promoting spread of wrong information. The authors have proposed a defense mechanism against these attacks that is based upon Bayesian Model. The proposed defense mechanism provides protection against these attacks so that fake or wrong information could be prevented. The authors evaluated the proposed mechanism in three different scenarios and presented the results in terms of pre- cision, recall and rate of change in rank. The results reveal almost 88%prevention against these attacks in comparison to the baseline systems.