The improved fuzzy clustering algorithm based on AFS theory and its applications to Wisconsin breast cancer data

In this paper, the AFS fuzzy logic clustering algorithm proposed by X.D. Liu has been studied further by the improvement of the algorithm. Instead of examples of less than 10 samples in Liu´s paper, we apply the improved algorithm to Wisconsin breast cancer data which has 699 samples and just the order relationships of the samples on each feature are used in the algorithm. This study shows that the AFS fuzzy logic clustering algorithm can obtain a high clustering accuracy based on the order relations on the features can compare with some classifiers.