Dereceli Bulanık Sistem Modelleri

"There is no claim that is certain...
but that doesn't stop us from making new claims."
Carneades, BC. 214-129

prof. Dr. I. A Masterpiece on Graded (Fuzzy) System Models by Burhan Türkşen!
This book focuses on new fuzzy system modeling approaches. These new approaches use type-1 and type-2 fuzzy sets and other flexible computational methods such as genetic algorithms and support vector machines for parameter optimization, membership function identification, indefinite range capture, and local function convergence. Two different versions of a new fuzzy clustering method with validity criteria are presented. The new methods presented in this book are listed in detail below:

• Two newly developed fuzzy clustering methods for regression and classification problem areas,
• A new type-1 developed fuzzy functions approach instead of fuzzy rule bases,
• A new type-2 developed fuzzy functions approach for uncertainty modeling,
• Extensions of type-1 and type-2 fuzzy functions for classification problem areas,
• Two new hybrid type-1 and type-2 fuzzy functions approaches using evolutionary computation methods,
• Extensions of hybrid approaches for classification problem domains,
• Two new clustering validity indexes to be used in the evaluation of newly developed fuzzy clustering methods for regression and classification problem areas,
• A new semi-parametric state-based inference method for type-1 enhanced fuzzy functions,
• A new state-based type reduction method for type-2 enhanced fuzzy functions,
• Extensions of Type-1 and Type-2 enhanced fuzzy functions and hybrid type-1 and type-2 enhanced fuzzy functions that use the classical fuzzy c-means clustering method instead of the improved fuzzy clustering method,
• A new performance measure used specifically for stock price prediction systems.
Since the word fuzzy was first used by LA Zadeh in 1965, it can sometimes cause controversy or misunderstanding. For Professor Zadeh's fuzzy term, "Well thought through, this term is as ironic as it is important." This expression, which we translated into Turkish as fuzzy, started to be better understood and found its true meaning in our language. In this book, instead of Fuzzy Logic, the term Gradual Logic is used from time to time. This is because, on the contrary to creating confusion in the use of the word fuzzy, it is to ensure that the word fuzzy can be better formed in the minds of the readers who will recognize this idea for the first time.

• Ordered (Fuzzy) Sets and Systems
• Type-1 Ordered Sets and Ordered Logic
• Type-2 Ordered Clusters
• Improved Fuzzy Clustering
• Fuzzy Clustering Algorithms
• Enhanced Fuzzy Clustering Algorithm (GBÖ - IFC)
• Two New Validity Indices for GBS and SDQ
• Ordered Regression Function Structures Approach
• Type-1 Ordered Regression Functions Approach Using Bco (T1BRF)
• Type-1 Enhanced Ordered Regression Functions Approach using GBI (T1GBRF)
• Modeling Uncertainty with Enhanced Ordered Functions
• Split Range Type-2 Advanced Order Functions
• Advantages of Uncertainty Modeling
• Experiments
• Analysis of Experiments
• All Discussions in Experiments