RT info:eu-repo/semantics/bookPart
T1 Robust Fuzzy Clustering via Trimming and Constraints
A1 Dotto, Francesco
A1 Farcomeni, Alessio
A1 García Escudero, Luis Ángel
A1 Mayo Iscar, Agustín
K1 Statistics
AB A methodology for robust fuzzy clustering is proposed. Thismethodology can be widely applied in very different statistical problems giventhat it is based on probability likelihoods. Robustness is achieved by trimminga fixed proportion of “most outlying” observations which are indeedself-determined by the data set at hand. Constraints on the clusters’ scattersare also needed to get mathematically well-defined problems and to avoid thedetection of non-interesting spurious clusters. The main lines for computationallyfeasible algorithms are provided and some simple guidelines abouthow to choose tuning parameters are briefly outlined. The proposed methodologyis illustrated through two applications. The first one is aimed at heterogeneouslyclustering under multivariate normal assumptions and the secondone migh be useful in fuzzy clusterwise linear regression problems.
PB Springer International Publishing
SN 978-3-319-42972-4
YR 2016
FD 2016
LK http://uvadoc.uva.es/handle/10324/21842
UL http://uvadoc.uva.es/handle/10324/21842
LA eng
NO Soft Methods for Data Science.  Editors:  Maria Brigida Ferraro, Paolo Giordani , Barbara Vantaggi , Marek Gagolewski , María Ángeles Gil, Przemysław Grzegorzewski , Olgierd Hryniewicz , Springer International Publishing, 2016, p. 197-204 (Advances in Intelligent Systems and Computing, 456)
NO Producción Científica
DS UVaDOC
RD 17-may-2024