POTCLAS - A FUZZY EXPERT-SYSTEM FOR THE CLASSIFICATION OF ARCHAEOLOGICAL POTTERY FRAGMENTS

The features of POTCLAS, a fuzzy expert system that aims at classifyin g archaeological pottery fragments, are illustrated in this paper. Fuz zy logic has allowed us to develop appropriate models to manipulate li nguistic variables and inferential mechanisms that are particularly su itable for the archaeological activity, characterized by the expert's remarks that are necessarily vague and imprecise. In particular, the r ole played by a suitable generalization of the rule ''Modus ponens'' i s emphasized and the PROLOG implementation of the system is outlined.