ABSTRACT
A common inference task consists of making a discrete prediction about some object given other details about the object. For instance, in financial credit assessment as discussed by Carter and Catlett (1987) we wish to decide whether to accept or reject a customer's application for a loan given particular personal information. This prediction problem is the basic task of many expert systems, and is referred to in artificial intelligence as the classification problem (where the prediction is referred to as the classification). The task is to learn a classifier given a training sample of classified examples. In credit assessment, classes are 'accept' and 'reject' (the credit application). A training sample in this case would be historical records of previous loan applications together with whether the loan went bad. This generic learning task is referred to as supervised learning in pattern recognition, and induction or empirical learning in machine learning (see, for instance, Quinlan, 1986). The statistical community uses techniques such as discriminant analysis and nearest neighbour methods, described, for instance, by Ripley (1987).
