nonparametric_naive_bayes {naivebayes}

Non-Parametric Naive Bayes Classifier

Description

nonparametric_naive_bayes is used to fit the Non-Parametric Naive Bayes model in which all class conditional distributions are non-parametrically estimated using kernel density estimator and are assumed to be independent.

Usage

nonparametric_naive_bayes(x, y, prior = NULL, ...)

Arguments

x

matrix with metric predictors (only numeric matrix accepted).

y

class vector (character/factor/logical).

prior

vector with prior probabilities of the classes. If unspecified, the class proportions for the training set are used. If present, the probabilities should be specified in the order of the factor levels.

...

other parameters to density (for instance adjust, kernel or bw).

Details

This is a specialized version of the Naive Bayes classifier, in which all features take on real values (numeric/integer) and class conditional probabilities are estimated in a non-parametric way with the kernel density estimator (KDE). By default Gaussian kernel is used and the smoothing bandwidth is selected according to the Silverman's 'rule of thumb'. For more details, please see the references and the documentation of density and bw.nrd0.

The Non-Parametric Naive Bayes is available in both, naive_bayes() and nonparametric_naive_bayes(). The latter does not provide a substantial speed up over the general naive_bayes() function but it is meant to be more transparent and user friendly.

The nonparametric_naive_bayes and naive_bayes() are equivalent when the latter is used with usekernel = TRUE and usepoisson = FALSE; and a matrix/data.frame contains only numeric variables.

The missing values (NAs) are omitted during the estimation process. Also, the corresponding predict function excludes all NAs from the calculation of posterior probabilities (an informative warning is always given).

Value

nonparametric_naive_bayes returns an object of class "nonparametric_naive_bayes" which is a list with following components:

data

list with two components: x (matrix with predictors) and y (class variable).

levels

character vector with values of the class variable.

dens

nested list containing density objects for each feature and class.

prior

numeric vector with prior probabilities.

call

the call that produced this object.

Author(s)

Michal Majka, [e-mail]

References

Silverman, B. W. (1986). Density Estimation for Statistics and Data Analysis. Chapman & Hall.

See Also

predict.nonparametric_naive_bayes, plot.nonparametric_naive_bayes, tables, get_cond_dist, %class%

Examples

# library(naivebayes)
data(iris)
y <- iris[[5]]
M <- as.matrix(iris[-5])

### Train the Non-Parametric Naive Bayes
nnb <- nonparametric_naive_bayes(x = M, y = y)
summary(nnb)
head(predict(nnb, newdata = M, type = "prob"))

###  Equivalent calculation with general naive_bayes function:
nb <- naive_bayes(M, y, usekernel = TRUE)
summary(nb)
head(predict(nb, type = "prob"))

### Change kernel
nnb_kernel <- nonparametric_naive_bayes(x = M, y = y, kernel = "biweight")
plot(nnb_kernel, 1, prob = "conditional")

### Adjust bandwidth
nnb_adjust <- nonparametric_naive_bayes(M, y, adjust = 1.5)
plot(nnb_adjust, 1, prob = "conditional")

### Change bandwidth selector
nnb_bw <- nonparametric_naive_bayes(M, y, bw = "SJ")
plot(nnb_bw, 1, prob = "conditional")

### Obtain tables with conditional densities
# tables(nnb, which = 1)