Users’ Internet Searches as Proxies for Disease Escalation Trends

Viral Hepatitis diseases are of the most infectious diseases in the world with 10s of millions. Over half the world’s population is exposed to the different hepatotropic viruses 1. In this research we studied the ability of Internet based search and keywords’ surveillance to correlate with infectious diseases escalation. With focus on USA, we collected data from CDC on Hepatitis diseases for several years and built a dataset of Internet search terms that can correlate with the volumes of reported cases of those diseases. We presented the final product as “best set of keywords” that can be used to predict future possible breakouts in Hepatitis. Linear regressions and decision trees were used to test the level of accuracy for the prediction based on Hepatitis search keywords