Generalized noise clustering based on non-Euclidean distance

Based on a new distance, Dave's generalized noise clustering (GNC) algorithm is extended to non-Euclidean generalized noise clustering (NGNC) model. Different from fuzzy c-means (FCM) model and GNC model which are based on Euclidean distance, the presented model is based on non-Euclidean distance. Based on robust statistical point of view and influence function, the non-Euclidean distance is more robust than the Euclidean distance. So the NGNC algorithm is more robust than the GNC algorithm. Moreover, with the new distance NGNC can deal with noises or outliers better than GNC and fuzzy c-means (FCM). The experimental results show the better performance of NGNC algorithm.