Web monitoring of emerging animal infectious diseases integrated in the French Animal Health Epidemic Intelligence System

摘要: Since 2013, the French Animal Health Epidemic Intelligence System (in French: Veille Sanitaire Internationale, VSI) has been monitoring signals of emergence new and exotic animal infectious diseases worldwide. Once detected, VSI team verifies issues early warning reports to health authorities when potential threats France are detected. To improve detection from online news sources, we designed Platform for Automated extraction Disease Information web (PADI-web). PADI-web automatically collects, processes extracts English-language epidemiological information Google News. The core component is a combined (IE) method founded on rule-based systems data mining techniques. IE approach allows key diseases, locations, dates, hosts number cases mentioned in news. We evaluated dataset 352 disease-related mentioning involved, cases. accurately identified (F-score) 95% hosts, respectively, 85% cases, 83% dates 80% locations assessed sensitivity detect primary outbreaks four emerging notifiable World Organisation (OIE). From January June 2016, detected 64% all African swine fever, 53% avian influenza, 25% bluetongue 19% foot-and-mouth disease. timely influenza disease Asia, i.e. they were 8 3 days before immediate notification OIE, respectively.