[A method to select diagnostic criteria from Bayes' theorem and information theory (author's transl)]

An algorithm is proposed to order a list of characteristics according to their efficiency for diagnosis. Two diagnostic ability coefficients (DAC), INF and TAX, are described. The calculus of INF coefficient is based on Bayes' theorem, using the occurrence of various diagnosis, and on information theory. TAX coefficient is a linear approach of the former. Both coefficients are tested with the study of 1891 strains of Enterobacteriaceae, originating from urines. The results are compared with those given by Gyllenberg's separation and characterization figures. Theoretical examples are also provided for computing purposes.