Finally, methods used in deception detection in social media must account for features of social context (such as friends and family of an individual) that have been found to increase the accuracy of detection of deception.18 The downside is social network analyses (SNAs) tend to be dramatically more expensive as networks grow. Simple SNA metrics (such as “betweeness centrality”) become overwhelmingly difficult to compute as networks grow (O(N3)) where N is the number of nodes and more advanced statistical methods (such as exponential random graph models using Markov chain Monte Carlo algorithms) are costly to compute. However, the potential for this newly available social data is apparent, and computational efficiency must be addressed in large social networks. On a positive note, one online trend is formation of small social networking sites5 and communities for which deception-detection methods may be more computationally feasible