Convert multiple binary columns to single categorical column [duplicate]

Question

This question already has an answer here:

• For each row return the column name of the largest value 7 answers

I have a table full of binary variables that I would like to condense down to categorical variables.

Very simplistically, I have is a data frame like this:

data <- data.frame(id=c(1,2,3,4,5,6,7,8,9), red=c("1","0","0","0","1","0","0","0","0"),blue=c("0","1","1","1","0","1","1","1","0"),yellow=c("0","0","0","0","0","0","0","0","1"))
data
  id   red   blue  yellow
1  1   1    0      0
2  2   0    1      0
3  3   0    1      0
4  4   0    1      0
5  5   1    0      0
6  6   0    1      0
7  7   0    1      0
8  8   0    1      0
9  9   0    0      1

And what I would like to get back would be:

  id   color 
1  1   red    
2  2   blue   
3  3   blue    
4  4   blue    
5  5   red    
6  6   blue    
7  7   blue    
8  8   blue    
9  9   yellow 

I hope there's a really simple answer for this.

Answer 1

You can get the values by making use of the column names and as.logical. However, since your "binary" columns are factors, you need to go though a few more hoops:

> apply(data[-1], 1, function(x) names(x)[as.logical(as.numeric(as.character(x)))])
[1] "red"    "blue"   "blue"   "blue"   "red"    "blue"   "blue"   "blue"   "yellow"

Bind this back with the first column (data[1]) to get the output you want.

cbind(data[1], 
      color = apply(data[-1], 1, 
                    function(x) names(x)[as.logical(as.numeric(
                      as.character(x)))]))
#   id  color
# 1  1    red
# 2  2   blue
# 3  3   blue
# 4  4   blue
# 5  5    red
# 6  6   blue
# 7  7   blue
# 8  8   blue
# 9  9 yellow

Alternatively, you can try the following:

data[-1] <- lapply(data[-1], function(x) as.numeric(as.character(x)))
temp <- subset(cbind(data[1], stack(data[-1])), values == 1, c("id", "ind"))
temp[order(temp$id), ]

Or, you can use a combination of "dplyr" and "tidyr", like this:

library(dplyr)
library(tidyr)

data %>%
  group_by(id) %>%
  mutate_each(funs(an = as.numeric(as.character(.)))) %>%
  gather(color, val, -id) %>%
  filter(val == 1) %>%
  select(-val) %>%
  arrange(id)
# Source: local data frame [9 x 2]
# 
#   id  color
# 1  1    red
# 2  2   blue
# 3  3   blue
# 4  4   blue
# 5  5    red
# 6  6   blue
# 7  7   blue
# 8  8   blue
# 9  9 yellow

Answer 2

Here's a simple base R vectorized solution using max.col

cbind(data[1L], color = names(data[-1L])[max.col(data[-1L] == 1L)])
#   id  color
# 1  1    red
# 2  2   blue
# 3  3   blue
# 4  4   blue
# 5  5    red
# 6  6   blue
# 7  7   blue
# 8  8   blue
# 9  9 yellow