In AnderEhu/sme:

library(SME)
library(knitr)

discretize

Discretization is the process of transforming a continuous-valued variable into a discrete one by creating a set of contiguous intervals (or equivalently a set of cutpoints) that spans the range of the variable's values.

There are diferent discretization methods, for example:

Equal interval width discretization: all bins have equal width.
Equal frequency discretization: all bins contain the same number of numerical values.
k-means discretization: consists of applying the classical K-means clustering to one-dimensional continuous data, k is equal to the number of bins.

Usage

discretize(x, method = "frequency", num.bins)

Arguments

x a numeric vector, matrix or dataframe (continuous variables)
method discretization method. Available are: "interval" (equal interval width), "frequency" (equal frequency), "cluster" (k-means clustering).
num.bins number of intervals.

Value

A discretized vector or dataframe

Examples

v.example.1 <- c(11.5, 10.2, 1.2, 0.5, 5.3, 20.5, 8.4)
v.example.2 <- c(0,4,12,16,16,18,24,26,28)
df.example.3 <- data.frame(c(0,4,12), c(16,16,18), c(24,26,28))
matrix.example.4 <- matrix(c(0,4,12,16,16,18,24,26,28), 3, 3, byrow = FALSE )

ew.discretize.example1 <- discretize(v.example.1, method="interval", 4)
print(ew.discretize.example1)

ew.discretize.example2 <- discretize(v.example.2, method="interval", 3)
print(ew.discretize.example2)

ew.discretize.example3 <- discretize(df.example.3, method="interval", 3)
kable(ew.discretize.example3)

ew.discretize.example4 <- discretize(matrix.example.4, method="interval", 2)
kable(ew.discretize.example4)

ef.discretize.example1 <- discretize(v.example.1, method="frequency", 4)
print(ef.discretize.example1)

ef.discretize.example2 <- discretize(v.example.2, method="frequency", 3)
print(ef.discretize.example2)

ef.discretize.example3 <- discretize(df.example.3, method="frequency", 3)
kable(ef.discretize.example3)

ef.discretize.example4 <- discretize(matrix.example.4, method="frequency", 2)
kable(ef.discretize.example4)


clustering.discretize.example1 <- discretize(v.example.1, method="clustering", 4)
print(clustering.discretize.example1)

clustering.discretize.example2 <- discretize(v.example.2, method="clustering", 3)
print(clustering.discretize.example2)

clustering.discretize.example3 <- discretize(df.example.3, method="clustering", 3) 
kable(clustering.discretize.example3)

clustering.discretize.example4 <- discretize(matrix.example.4, method="clustering", 2)   
kable(clustering.discretize.example4)