Measuring population health using personal health messages.

摘要: Personal health messages – inter patient communications within online communities; represent a new path towards providing continuous information about derived status. We apply natural language processing techniques to personal from message boards demonstrate the ability track trends in people’s positive or negative opinion (sentiment) regarding particular drugs over time. The significant changes sentiment correspond FDA announcements and other publicity. envision such analysis as scalable tool for pharmacovigilance hypothesis generation possible adverse drug reactions.