TY - JOUR
T1 - A “Density-Based” Algorithm for Cluster Analysis Using Species Sampling Gaussian Mixture Models
AU - Argiento, Raffaele
AU - Cremaschi, Andrea
AU - Guglielmi, Alessandra
PY - 2014
Y1 - 2014
N2 - We propose a new model for cluster analysis in a Bayesian nonparametric framework.
Our model combines two ingredients, species sampling mixture models of Gaussian
distributions on one hand, and a deterministic clustering procedure (DBSCAN) on the
other. Here, two observations from the underlying species sampling mixture model
share the same cluster if the distance between the densities corresponding to their
latent parameters is smaller than a threshold; this yields a random partition which is
coarser than the one induced by the species sampling mixture. Since this procedure
depends on the value of the threshold, we suggest a strategy to fix it. In addition, we
discuss implementation and applications of the model; comparison with more standard
clustering algorithms will be given as well. Supplementary materials for the article are
available online.
AB - We propose a new model for cluster analysis in a Bayesian nonparametric framework.
Our model combines two ingredients, species sampling mixture models of Gaussian
distributions on one hand, and a deterministic clustering procedure (DBSCAN) on the
other. Here, two observations from the underlying species sampling mixture model
share the same cluster if the distance between the densities corresponding to their
latent parameters is smaller than a threshold; this yields a random partition which is
coarser than the one induced by the species sampling mixture. Since this procedure
depends on the value of the threshold, we suggest a strategy to fix it. In addition, we
discuss implementation and applications of the model; comparison with more standard
clustering algorithms will be given as well. Supplementary materials for the article are
available online.
KW - Bayesian nonparametrics
KW - DBSCAN algorithm
KW - Dirichlet process
KW - Bayesian nonparametrics
KW - DBSCAN algorithm
KW - Dirichlet process
UR - http://hdl.handle.net/10807/148064
UR - https://www.scopus.com/inward/record.uri?eid=2-s2.0-84908061167&doi=10.1080/10618600.2013.856796&partnerid=40&md5=58cac680a5f40d4b04e51e37ad727714
U2 - 10.1080/10618600.2013.856796
DO - 10.1080/10618600.2013.856796
M3 - Article
SN - 1061-8600
VL - 23
SP - 1126
EP - 1142
JO - Journal of Computational and Graphical Statistics
JF - Journal of Computational and Graphical Statistics
ER -