A Model of Case Based Reasoning Using Intuitionistic Fuzzy Sets

Real world decision making is rather complicated -either there are not adequate models or the models are too expensive to derive. The paradigm of case-based reasoning (CBR) of matching new problems to "cases" from a historical database and then adapting successful solutions available from the past to current situations seems to be useful. In this paper we propose to apply intuitionistic fuzzy sets in case-based reasoning. We address in details one but very important problem one inevitably meets when coping with the central tasks of CBR -the calculation of distances (that represent in our paper a similarity) between intuitionistic fuzzy sets. Distances are indispensable, e.g., when concluding about the similarity of cases. We give reasons for using an extended definition of distances for intuitionistic fuzzy sets. It means not only a correct approach from the point of view of fulfilling some mathematical properties but also from the point of view of effective and efficient decision making -i.e., the way making it possible to properly use all the relevant information available. We also discuss incorrect approaches leading to jumping to conclusions.