FCMdd-type linear fuzzy clustering for incomplete non-Euclidean relational data

A linear fuzzy clustering model based on Fuzzy c-Medoids (FCMdd) was proposed for extracting intrinsic local linear substructures from relational data. In the same way with Non-Euclidean Relational Fuzzy (NERF) c-Means, β-spread transformation was also proved to be useful for handling non Euclidean relational data in FCMdd-type linear clustering. In this paper, the FCMdd-type linear clustering model is further modified in order to handle incomplete data including missing values. In several numerical experiments, it is demonstrated that some pre-imputation strategies contribute to properly selecting representative medoids of each cluster.