Fuzzy Rule Generation

In recent years, a wide range of problems are handled by fuzzy models. Fuzzy model prefers on classic models due its ability to deal with imprecise data, model the complex nonlinear problems and acquiring knowledge with these models. Fuzzy rule is a special case of fuzzy modeling, where its working principle is representing the behavior of the system by a set fuzzy if-then classification rules. An intelligent technique has been proposed in this thesis which it depends on a Takagi- Sugeno-Kang (TSK) fuzzy modeling, subtractive clustering method and an efficient gradient descent algorithm. This approach uses subtractive clustering method to extract the fuzzy classification rules from data; the rule's parameters are then optimized by using an efficient gradient descent algorithm. The dataset firstly is divided into main classes, and then the subtractive algorithm is applied for each class. The clusters centers set, and sigma set are generated from this clustering process. To enhance the performance of the system, a gradient descent method is employed which represents the optimization method that it is used to adjust the value of the clusters centers and sigma values.