Exploiting similarity and experience in decision making

The idea of case-based decision making has recently been proposed as an alternative to the expected utility theory. A case-based decision maker learns by storing already experienced decision problems, along with a rating of the results. Whenever a new problem needs to be solved, possible actions are assessed on the basis of experience from similar situations in which these actions have already been applied. In this paper, we consider case-based decision making within the context of instance-based learning, which is a special type of machine learning method. From this consideration we suggest alternative case-based decision principles. These principles are motivated from a computational point of view and characterized axiomatically. Moreover, the possibility of applying case-based decision making in approximate reasoning is briefly discussed