mclust (version 5.4.7)

predict.MclustSSC: Classification of multivariate observations by semi-supervised Gaussian finite mixtures

Description

Classify multivariate observations based on Gaussian finite mixture models estimated by MclustSSC.

Usage

# S3 method for MclustSSC
predict(object, newdata, …)

Arguments

object

an object of class 'MclustSSC' resulting from a call to MclustSSC.

newdata

a data frame or matrix giving the data. If missing the train data obtained from the call to MclustSSC are classified.

…

further arguments passed to or from other methods.

Value

Returns a list of with the following components:

classification

a factor of predicted class labels for newdata.

z

a matrix whose [i,k]th entry is the probability that observation i in newdata belongs to the kth class.

See Also

MclustSSC.

Examples

# NOT RUN {
X <- iris[,1:4]
class <- iris$Species
# randomly remove class labels
set.seed(123)
class[sample(1:length(class), size = 120)] <- NA
table(class, useNA = "ifany")
clPairs(X, ifelse(is.na(class), 0, class),
        symbols = c(0, 16, 17, 18), colors = c("grey", 4, 2, 3),
        main = "Partially classified data")

# Fit semi-supervised classification model
mod_SSC  <- MclustSSC(X, class)

pred_SSC <- predict(mod_SSC)
table(Predicted = pred_SSC$classification, Actual = class, useNA = "ifany")

X_new = data.frame(Sepal.Length = c(5, 8),
                   Sepal.Width  = c(3.1, 4),
                   Petal.Length = c(2, 5),
                   Petal.Width  = c(0.5, 2))
predict(mod_SSC, newdata = X_new)
# }