Abstract:
Systems and methods for predicting virality of a content item are disclosed. A method includes: receiving a social network structure; identifying communities within the social network structure, where communities are identified as dense subnetworks in the social network structure; receiving social network content that includes one or more content items; and identifying one or more content items that are predicted to become viral based on utilization of the content items between different communities in the social network structure.