Abstract
Clustering is a technique of unsupervised classification. The methods are classified into two types, one is hierarchical and the other is non-hierarchical. Fuzzy Non-metric Model (FNM) is a method of non-hierarchical clustering. FNM is very useful because Euclidean spaceis not required on data space and only similarity or dissimilarity between data is needed. However FNM cannot handle uncertain data, e.g. incomplete data, or data which have errors. In order to handle such data, concept of data with tolerance has been proposed. The methods using the concept can handle the uncertain data in the framework of optimization, e.g. Fuzzy $c$-means for data with Tolerance (FCM-T). In this report, first we will propose new clustering algorithm to apply the concept of tolerance to FNM. Second, we will show that the proposed algorithm handle incomplete data. Third, we will verify the effectiveness of the proposed algorithm in comparison with other ones for incomplete data through some numerical examples.
