The Data Fusion for Hypothesis Comparison system provides a technology for generating knowledge from data and using it to compare competinghypotheses. Current approaches to data fusion in the aid of decision support systems cannot dynamically synthesize knowledge about data, and filter out the more from the less likely hypotheses that result. Further they provide no means to control data deluge, missing data, or adjust as unanticipated results are discovered. By utilizing recent advances in data mining, both deductive and inductive, problems in the current technologies can be overcome. The result is a system that can support commanders at all levels by providing the best hypothesis available and giving a means of varying the parameters associated with those recommendations.
Benefits: The benefits will accrue from improved decision making due to a better understanding of the hypotheses and assumptions used in building the decision model. Commercial applications will be found in all business areas where data from databases are used to support financial, marketing and operational decisions.
