Java Fuzzy Kit (JFK): A shell to build fuzzy inference systems according to the generalized principle of extension

摘要

In this article the author presents JFK, which stands for Java Fuzzy Kit. JFK is an Application Programming Interface (API) that complies with both, a general structure of a fuzzy rule base and the necessary processing to compute the generalized principle of extension. A recurrent structure is found for a class of fuzzy expert systems, known as the Mamdani model. This leads to claim that a design pattern exists, since core objects, which are present regardless the specific application are identified. However, there is not a general shell to build fuzzy expert systems, and this provokes that current fuzzy expert systems are build on an ad-hoc basis. The modelling of JFK is done according to the Unified Modelling Language specifications. Along with the UML modelling three important algorithms are described, which serve to perform the generalized principle of extension. The usage of JFK is illustrated with an example, namely the ranking of swimmers. Preliminary results on this study case are the basis to propose the realization of fuzzy distributed decision systems. This goal is accomplished by providing agents with fuzzy expert systems, with the integration of JFK and the standardized platform JADE (Java Agent Development Environment).