|
Abstract:
|
The prompt detection and forecasting of infectious diseases with rapid transmission and high virulence are critical in the effective defense against these diseases. Despite many promising approaches in modern surveillance methodology, the lack of observations for near real-time forecasting is still the key challenge obstructing operational disease prediction and control. In contrast, non-traditional data sources, such as online social media, create a new momentum for real-time epidemiological forecasting and have potential to revolutionize modern biosurveillance capabilities by predicting an event before its typical manifestation and before patient-healthcare interaction. In this talk we investigate utility of Twitter to serve as a proxy for yet unobserved or yet not publicly unavailable data on flu occurrence and propose a predictive platform for individual-specific disease dynamics by accounting for heterogeneous everyday social interactions, space-time, and socio-demographic information.
|
