A computerised system for the identification of lactic acid bacteria.

A generic computerised system for the identification of bacteria was developed. The system is equipped with a key to the identification of lactic acid bacteria. The identification is carried out in two steps. The first step distinguishes groups of bacteria by following a decision tree with general identification tests. The second step in the identification is the distinction of species within a group on the basis of biochemical fermentation patterns. During grouping, probabilities for test failure are used. These probabilities can be used for assessing the quality of a given test answer. The probabilities are also used to select the most probable test answer in cases where the test result is inconclusive. The probabilities of test failure were determined by a group of experts and a group of potential users of the identification system. During species identification, similarity indices are calculated for all bacteria in a group. The described identification system has the ability to "learn" from different sessions in the species identification step, improving both identification speed and accuracy. Because of the versatile way in which the system is set up, it can very easily be expanded with identification keys to other organisms.