TY - GEN
T1 - Group-wise penalized estimation schemes in model-based clustering = Strategie di stima penalizzata a livello di gruppo nel
clustering basato su modello
AU - Casa, A.
AU - Cappozzo, Andrea
AU - Fop, M.
PY - 2022
Y1 - 2022
N2 - Gaussian mixture models provide a probabilistically sound clustering approach.\r\nHowever, their tendency to be over-parameterized endangers their utility\r\nin high dimensions. To induce sparsity, penalized model-based clustering strategies\r\nhave been explored. Some of these approaches, exploiting the link between\r\nGaussian graphical models and mixtures, allow to handle large precision matrices,\r\nencoding variables relationships. By assuming similar components sparsity levels,\r\nthese methods fall short when the dependence structures are group-dependent. Our\r\nproposal, by penalizing group-specific transformations of the precision matrices, automatically\r\nhandles situations where under or over-connectivity between variables\r\nis witnessed. The performances of the method are shown via a real data experiment
AB - Gaussian mixture models provide a probabilistically sound clustering approach.\r\nHowever, their tendency to be over-parameterized endangers their utility\r\nin high dimensions. To induce sparsity, penalized model-based clustering strategies\r\nhave been explored. Some of these approaches, exploiting the link between\r\nGaussian graphical models and mixtures, allow to handle large precision matrices,\r\nencoding variables relationships. By assuming similar components sparsity levels,\r\nthese methods fall short when the dependence structures are group-dependent. Our\r\nproposal, by penalizing group-specific transformations of the precision matrices, automatically\r\nhandles situations where under or over-connectivity between variables\r\nis witnessed. The performances of the method are shown via a real data experiment
KW - EM algorithm
KW - Gaussian graphical models
KW - Graphical lasso
KW - Model-based clustering
KW - EM algorithm
KW - Gaussian graphical models
KW - Graphical lasso
KW - Model-based clustering
UR - https://publicatt.unicatt.it/handle/10807/306444
M3 - Conference contribution
SN - 9788891932310
SP - 534
EP - 539
BT - SIS 2022 | Book of Short Papers
PB - Pearson
ER -