On clustering based on homogeneity

The clustering technique in data analysis has had two main problems. One of them is how to determine the number of clusters and the other is concerned with the interpretation of the clustering result, i.e. what the obtained clusters mean. Fuzzy clustering has also had these problems. In this paper, we focus on the problems of fuzzy clustering. The merit of fuzzy clustering is that we can consider not only the status of belonging to the clusters but also how much the objects belong to the clusters. So, we can obtain the clustering result as the degree of belongingness of objects to the clusters, and these values are usually not discrete. Using this feature and the idea of homogeneity from homogeneity analysis, we propose a model to obtain an interpretation of the fuzzy clusters