The dynamic competition hypothesis

We study competitive and negative interactions in real world social network in which nodes represent agents and edges appear over discrete time step. We consider directed social network of competing agents that evolve dynamically over time, where directed edges represent some kind of negative relationships between the agents in the social network. We present a novel hypothesis to identify the alliances and leaders within the the dynamic competition networks. We verify our hypothesis by using historical voting data of the social game shows Survivor and Big Brother.