Charla CLUSTERING ANALYSIS (Mar 11 Dic, 15:00 - 16:30 h - Evento UCSP)

Inscripciones Eventos UCSP

Mar 11 Dic, 15:00 - 16:30 h

Expositor: Profesor James Bezdek 

(Conferencista distinguido para educación en países en desarrollo) B.S.C.E, University of Nevada (Reno) Ph.D., Cornell University Founding editor of International Journal of Approximate Reasoning and IEEE Transactions on Fuzzy Systems. Past president of three professional societies: North American Fuzzy Information Processing Society (NAFIPS), International Fuzzy Systems Association (IFSA) and IEEE Neural Networks Council. Fellow of the IEEE and IFSA. Recipient of the IEEE Millenium, Fuzzy Systems Pioneer and Frank Rosenblatt Medals. Resumen: The discussion begins by characterizing the three canonical problems of clustering: tendency assessment (does the data have cluster substructure?); clustering (how do we find partitions of the data?); and validation (are the partitions we find accurate and/or useful?). Then I discuss the four types of models used in clustering, and give a concrete example of each type. The four model types and the example algorithm discussed are: partition only (single linkage); prototype only (self-organizing maps); partition and prototypes (hard, fuzzy and possibilistic c-means); and (partition, prototype, other parameter) models (EM algorithm for Gaussian mixture decomposition). Fecha: Martes 11 de Diciembre Hora: 15:00 - 16:30 h Lugar: Auditorio Campiña Paisajista