Toward Predicting Collective Behavior via Social Dimension Extraction

The SocioDim framework demonstrates promising results toward predicting collective behavior. However, many challenges require further research. For example, networks in social media are continually evolving, with new members joining a network and new connections established between existing members each day. This dynamic nature of networks entails efficient update of the model for collective behavior prediction. It is also intriguing to consider temporal fluctuation into the problem of collective behavior prediction.