Department of Pattern Recognition

(eng): Considering different application possibilities of product distribution mixtures we have proposed three formal tools in the last years, which can be used to accumulate decision-making know-how from particular diagnostic cases. First, we have developed a structural mixture model to estimate multidimensional probability distributions from incomplete and possibly weighted data vectors. Second, we have shown that the estimated product mixture can be used as a knowledge base for the Probabilistic Expert System (PES) to infer conclusions from definite or even uncertain input information. Finally we have shown that, by using product mixtures, we can exactly optimize sequential decision-making by means of the Shannon formula of conditional informativity. We combine the above statistical tools in the framework of an interactive open-access medical diagnostic system with automatic accumulation of decision-making knowledge.