Deep Learning framework to model Influenza predictions using aggregated Google Search query data
Introduction
Influenza outbreaks cause up to 500,000 deaths a year worldwide, and an estimated 3,000–50,000 deaths a year in the United States. The ability to effectively prepare for and respond to outbreaks heavily relies on the availability of accurate real-time estimates and the existing methods remains limited. Traditional flu surveillance systems, such as Center for Disease Control and Prevention’s (CDC) influenza reports lag behind real-time by one to two weeks, whereas information contained in internet users’ search activity is available in near real-time. This project proposes and implements a framework for epidemiological predictions (such as influenza/flu prediction) using internet users’ search activity.
Problem Description
Building a real time framework using online activity data (like Google Search Trends) using Deep Learning and traditional forecasting methods and comparing the results with Google’s prediction of flu trends (GFT) in the United States.