Nowadays the amount of digitalized images is growing faster than ever. In this paper, we discuss the mechanisms of fuzzy clustering and fuzzy clustering with partial supervision in the analysis and classification of images. These approaches are strongly influenced by some parameters such as fuzzification coefficient and the percentage of labeled images. Aim of this paper is to propose a graphical representation of the inter-prototypes distance obtained using a multidimensional scaling method. The results demonstrate that this representation is strongly related with the obtained classification performance.
A Multidimensional Scaling Based GUI To Evaluate Partial Supervision Effects On Prototypes Spatial Localization In Fuzzy Clustering / Amato, Alberto; Di Lecce, Vincenzo; Pedrycz, Witold. - ELETTRONICO. - (2007), pp. 1179-1185. (Intervento presentato al convegno 12th International Conference on Fuzzy Theory & Technology, 2007 tenutosi a Utah - USA nel July 18-24-2007) [10.1142/9789812709677_0167].
A Multidimensional Scaling Based GUI To Evaluate Partial Supervision Effects On Prototypes Spatial Localization In Fuzzy Clustering
Alberto Amato;Vincenzo Di Lecce;
2007-01-01
Abstract
Nowadays the amount of digitalized images is growing faster than ever. In this paper, we discuss the mechanisms of fuzzy clustering and fuzzy clustering with partial supervision in the analysis and classification of images. These approaches are strongly influenced by some parameters such as fuzzification coefficient and the percentage of labeled images. Aim of this paper is to propose a graphical representation of the inter-prototypes distance obtained using a multidimensional scaling method. The results demonstrate that this representation is strongly related with the obtained classification performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
