Modelling how information flows over social networks is fundamental to understanding important phenomena such as memes, vitality, and social influence. This project combines new mathematical techniques for estimating social information flow and predictability with large-scale data science methods to better understand the information dynamics within social networks.