This function can be used for DBN-based clustering. It is the same function as bnclustOmics, but it also works for time series data.
clustDBN(
dbndata,
staticnodes = 0,
blacklist = NULL,
edgepmat = NULL,
kclust = 2,
chixi = 0.5,
seed = 100,
err = 1e-06,
maxEM = 10,
hardlim = 6,
deltahl = 2,
nit = 5,
epmatrix = TRUE,
plus1it = 4,
nruns = 1,
startpoint = "mclustPCA",
baseprob = 0.4,
commonspace = TRUE,
verbose = TRUE,
samestruct = TRUE,
pickmax = TRUE
)object of class 'bnclustOmics' containing the results of Bayesian-network based clustering: cluster assignments, networks representing the clusters
data matrix; rows are observations, columns are variables; static nodes have to be in the first column of the data
(integer) number of static nodes in a DBN
adjacency matrix containing information about which edges will be blacklisted in structure search
penalization matrix of the edges in structure learning
the number of clusters (mixture components)
prior pseudocounts used for computing parameters for binary nodes
integer number set for reproducibility
convergence criteria
maximum number of EM iterations (structural)
maximum number of parents per node when learning networks
additional number of parents when sampling from the common search space
number of internal iteration in structural EM
(logical) indicates if the matrices containing posterioir probabilities of single edges should be returned
maximum number of search space expansion iterations when performing structure search
number of runs of the EM algorithm
defines which algorithm is used to define starting cluster memberships: possible values "random", "mclustPCA" and "mclust"
defines the base probability of cluster membership when "mclustPCA" or "mclust" used as starting point
(logical) defines if the sampling has to be performed from the common search space
defines if the output messages should be printed
(logical) defines if initial and intrinsic part of transition structures should be the same
(logical) if TRUE only maximum EM run is returned
Polina Suter