Performance Comparison of Rule Generation Method Substractive Clustering and Fuzzy C-Means Clustering on Sugeno's Inference for Stroke Risk Detection

Abstract

Abstract - Fuzzy Inference is one method that can
solve the problem of uncertainty in a decision-making
or classification well. In inference, fuzzy rules that
represent the need of expert knowledge in the relevant
fields, so that the classification given decision or be
appropriate expert knowledge. However there are times
when experts are less able to represent the rules of the
appropriate knowledge or knowledge that there is need
of too many rules, so we need a method that can
generate rules based on the data given expert.
At issue troke s disease risk detection, it also occurs
because of the research that has been done by taking the
direct rule of experts, it turns out less than the maximum
accuracy, still 82.89%. Substractive methods
Clustering and Fuzzy C-Means (FCM) could generate
rules by grouping algorithm, in which the existing
training data are grouped in common and the rules of
the group raised. Differences in the two methods are in
determining the center of the cluster and assign each
incoming data which groups.
Based on research that has been done, substractive
average Clustering membrika better accuracy is
84.46%, while 73.81% FCM. However, in the
processing time FCM faster at 16.75 seconds to give an
average processing time of 13:02 seconds.