ABSTRACT
Epidemiological models have been extremely effective in public health for both analysis and policy making. They formalize insights into biological and social processes of disease propagation into (low parameter) mathematical models. The work reported here applied relatively simple epidemiological models to the spread of ideas extracted by applying statistical language processing methods to real-world data. These models reproduced observed patterns of information spread, demonstrating the applicability and utility of epidemiological models. Both epidemiology and statistical language processing offer methods that can enhance our results. Such approaches can be applied to track, to evaluate, and ultimately to predict, spread of ideas across multiple temporal and geographic scales, in a variety of media, with widespread military, intelligence, and other applications.
