Count unique values of a column by pairwise combinations of another column in R

Question

Let\'s say I have the following data frame:

   ID Code
1   1    A
2   1    B
3   1    C
4   2    B
5   2    C
6   2    D
7   3    C
8   3    A
9   3    D
10          


        

Answer 1

Using base only:

df <- data.frame(ID=c(1,1,1,2,2,2,3,3,3,3,4,4), 
                 code=c("A", "B", "C", "B", "C", "D", "C", "A", "D", "B", "D", "B"), stringsAsFactors =FALSE)
# Create data.frame of unique combinations of codes
e <- expand.grid(df$code, df$code)
e <- e[e[,1]!=e[,2],]
e1 <- as.data.frame(unique(t(apply(e, 1, sort))), stringsAsFactors = FALSE)

# Count the occurrence of each code combination across IDs
e1$count <- apply(e1, 1, function(y) 
                  sum(sapply(unique(df$ID), function(x) 
                             sum(y[1] %in% df$code[df$ID==x] & y[2] %in% df$code[df$ID==x]))))

# Turn the codes into a string and print output
out <- data.frame(Code.Combinations=do.call(paste, c(e1[,1:2], sep=", ")),
                  Count.of.ID=e1$count, stringsAsFactors = FALSE)


out
#   Code.Combinations Count.of.ID
# 1              A, B           2
# 2              A, C           2
# 3              A, D           1
# 4              B, C           3
# 5              B, D           3
# 6              C, D           2