Chattergrabber is an open source, natural language processing based toolkit for public health social media surveillance. ChatterGrabber is designed to collect and categorize a high volume of content at a low cost, providing a readily deployable solution for Epidemiologists to track emergent outbreaks in the field and an additional signal for syndromic surveillance. Sensitivity and specificity of results are maximized through the use of a novel pull method and genetic algorithm optimization of text classifiers. This enables the creation of long term surveillance experiments wholly independent of member reporter networks and hashtag tracking. Such records may also yield additional soft data through shared symptoms, rumors, and observations crucial to an epidemiological investigation.