Fuzzy Rule Generation for Diagnosis of Coronary Heart Disease Risk Using Substractive Clustering Method

摘要

Fuzzy modeling techniques have been widely used to solve the uncertainty problems. A diagnosis of coronary heart disease (CHD) consists of some parameters numerical value of lingustics data. It can be implemented using fuzzy system through construction of the rules which relate to the data. However, the range of linguistics value is determined by an expert that depends on his knowledge to interpret the problem. Therefore, we propose to generate the rules automatically from the data collection using subtractive clustering and fuzzy inference Tagaki Sugeno Kang orde-1 method. The subtractive clustering method is a clustering algorithm to look for data clusters that serve as the fuzzy rules for diagnosis of CHD risk. The selected cluster number is determined based on the value of variant boundaries. Hence, it is applied to fuzzy inference system method, Takagi Sugeno Kang order-1, which determines diagnnosis of the desease. The advantage of this method is applicable to generate the fuzzy rules without defining and describing from an expert.