Rule-based classification procedures related to the unprecisely formulated expert rules

The paper is devoted to the pattern recognition procedure based on the set of expert rules with unprecisely formulated weights understood as conditional probabilities. Adopting the probabilistic model, the recognition algorithm is derived and evaluation of its probability of misclassification is given. Furthermore, the case with both expert rules and the learning set is considered. The proposed algorithms were empirically tested on the computer generated data and in computer aided diagnosis of acute renal failure and compared with sample based (k-NN) algorithm.