I think that problem is a non-linear one. I'm pretty sure that it's not possible to predict the behavioral outcomes from the reactions of emotional states. The

I think that problem is a non-linear one. I'm pretty sure that it's not possible to predict the behavioral outcomes from the reactions of emotional states. The science only goes so far as to provide a proof that it's possible to polarize sentiment by identifying the community bridge in a network. That bridge would be the user that has optimal influence in both groups.