The Social Media Genome: Modeling Individual Topic-Specific Behavior in Social Media

摘要

Information propagation in social media depends not only on the staticfollower structure but also on the topic-specific user behavior. Hence novelmodels incorporating dynamic user behavior are needed. To this end, we proposea model for individual social media users, termed a genotype. The genotype is aper-topic summary of a user's interest, activity and susceptibility to adoptnew information. We demonstrate that user genotypes remain invariant within atopic by adopting them for classification of new information spread inlarge-scale real networks. Furthermore, we extract topic-specific influencebackbone structures based on information adoption and show that they differsignificantly from the static follower network. When employed for influenceprediction of new content spread, our genotype model and influence backbonesenable more than $20% improvement, compared to purely structural features. Wealso demonstrate that knowledge of user genotypes and influence backbones allowfor the design of effective strategies for latency minimization oftopic-specific information spread.