Incorporating Prior Probabilities and Misclassification Costs into Network Training: An Example from Medical Prognosis.

摘要

Feed-forward layered networks trained on a pattern classification task in which the number of training patterns in each class in nonuniform, bias strongly in favour of those classes with largest membership. This is an unfortunate property of networks when the relative importance of classes with smaller membership is much greater than that of classes with many training patterns. In addition, there are many pattern classification tasks where different penalties are associated with misclassifying a pattern belonging to one class as another class. It is not generally known how to compensate for such effects in network training. This paper discusses an analytical regularisation scheme whereby prior expectations of class importance occurring in the generalisation data and misclassification costs may be incorporated into the training phase, thus compensating for the uneven and unfair class distributions occurring in the training set. The effects of the proposed scheme on the feature extraction criterion employed in the hidden layer of the network is discussed. An illustration of the results is presented by considering a real medical prognosis problem concerning data collected from head-injured coma patients. Relationships between least mean square error minimisation and Bayesian minimum risk estimation is mentioned and the importance and relevance of input/output coding schemes for network performance is considered. Keywords: Computer aided diagnosis, Medical computer applications, Artificial intelligence, Great Britain. (kt)