Comprehensive forecasting and prediction of events through textual, semantic, and visual analytics of crowdsourced data

Abstract

In this research we propose to bring our (UML and USP) combined experience and strength in the sciences of Visual Analytics, network visualization and monitoring, data mining, mobile and cloud computing, and the Semantic Web together to develop (and, in the future, deploy) an analytics infrastructure that will gather information from multiple users' sources, such as mobile phones, social networks, Web site traffic and the like, and will provide visual analytics to assist the detection and prediction of an onslaught of a threatening event, whether within public health, homeland security, natural disasters, or any other hard-to-predict events that might have a negative impact on populations, but can be mitigated with enough a priori alert. The same tools could be utilized to predict expected demand for certain products and services, and facilitate supply logistics to respond to such demand. (AU)