Dummify Variables in R

For Machine Learning we often need to create dummy variables for categorical data.

Caret package of R has dummyVars function to dummify the variables.

Eg. lets say we have a data frame (df) as following
df <- data.frame(age=c(40,47,34,21,34,67,56,78),
                 gender=c("M","M","F","F","M","F","M","F"),
                 favdrink=c("Coffee","Tea","Cold-drink","Tea","Coffee","Tea","Coffee","Cold-drink")
                )
print(df)
  age gender   favdrink
  40      M     Coffee
  47      M        Tea
  34      F Cold-drink
  21      F        Tea
  34      M     Coffee
  67      F        Tea
  56      M     Coffee
  78      F Cold-drink

library(caret)

# dummify the data
# dummyVars function breaks out unique values from a column into individual columns 
 
# Lets dummify gender column
dummy1 <- dummyVars("~ gender", data=df, fullRank = T)
trsf <- data.frame(predict(dummy1, newdata = df))
print(trsf)
  gender.M
        1
        1
        0
        0
        1
        0
        1
        0
dummy Variable gender.M 1 means male, 0 means female

We can dummify all or more than 1 variable as following

dummy2 <- dummyVars("~ gender + favdrink", data=df, fullRank = T)
trsf <- data.frame(predict(dummy2, newdata = df))
print(trsf)
  gender.M favdrink.Cold.drink favdrink.Tea
        1                   0            0
        1                   0            1
        0                   1            0
        0                   0            1
        1                   0            0
        0                   0            1
        1                   0            0
        0                   1            0
 
if favdrink.Cold-drink and favdrink.Tea are both zero, it means person has favdrink
as Coffee for this dataset