A Scalable Data Mining Approach for Providing Public Health with Disease Incidence Predictions Weeks in Advance

735 he Johns Hopkins University Applied Physics Laboratory (APL) has developed a novel and scalable data mining and fuzzy association rule-making approach to deriving disease incidence predictions several weeks in advance of an outbreak. This capability provides a new set of information that may be used by decision makers in conjunction with other complementary information about the country (e.g., infrastructure, disease history, agriculture, and U.S. and local military and civilian populations) from a variety of other sources (e.g., intelligence and disease experts). The prediction of the future infectious disease incidence provides the decision maker with enhanced ability to determine whether to enable deployment of measures to increase and focus biosurveillance and/or to plan and enable mitigation efforts to reduce morbidity and mortality well in advance of the start of the outbreak. A Scalable Data Mining Approach for Providing Public Health with Disease Incidence Predictions Weeks in Advance