Richard A. Caruana

Learn More
A surge of development of new public health surveillance systems designed to provide more timely detection of outbreaks suggests that public health has a new requirement: extreme timeliness of detection. The authors review previous work relevant to measuring timeliness and to defining timeliness requirements. Using signal detection theory and decision(More)
The recent series of anthrax attacks has reinforced the importance of biosurveillance systems for the timely detection of epidemics. This paper describes a statistical framework for monitoring grocery data to detect a large-scale but localized bioterrorism attack. Our system illustrates the potential of data sources that may be more timely than traditional(More)
The use of numerical uncertainty representations allows better modeling of some aspects of human evidential reasoning. It also makes knowledge acquisition and system development, test, and modification more difficult. We propose that where possible, the assignment and/or refinement of rule weights should be performed automatically. We present one approach(More)
Preliminary results on seventeen test problems have shown the mdpGA to be able to solve problems more efficiently than a simple parallel genetic algorithm. In order to further evaluate the worth of this structure, both harder test problems and different population topologies should be explored. Perhaps the most pressing topic for future research is the need(More)
The standard methodology in connectionism is to break hard problems into simpler, reasonably independent subproblems, learn the subproblems separately, and then recombine the learned pieces 15]. This modularization can be counterproductive because it eliminates a potentially critical source of inductive bias: the inductive bias inherent in the similarity(More)
Preliminary results on seventeen test problems have shown the mdpGA to be able to solve problems more efficiently than a simple parallel genetic algorithm. In order to further evaluate the worth of this structure, both harder test problems and different population topologies should be explored. Perhaps the most pressing topic for future research is the need(More)
This thesis examines nonlinear axis scaling and its impact on the modeling of inter-attribute relationships. Through automated methods, the described system identifies possible scaling methods; decides which attributes serve as inputs or outputs; and builds regression trees that quantify these relationships. While the experiments focus on the accuracy and(More)