CODEWITHSUNDEEP
rlistmatrixvector
El David
El David

Reputation: 395

comparative matrix into percentage in R

I have created a matrix that compares strings of vectors within a list to each other. 

sapply(names(setlist), function(x) sapply(names(setlist), function(y) sum(setlist[[x]] %in% setlist[[y]])))


              A             B             C              D 
A             50            1             0              0   
B             1             6             0              0
C             0             0             51             8
D             0             0             8              46

For example the number of strings within vector A and vector B that are the exact same are 1, with 50 total strings in A and 6 within B.

I would like to normalize the data so that, using the example above. The total stings of A and B is 56, so divide 1 by 56 = .018. The end result should look something like this:

              A             B             C              D 
A             .5            .018          0              0   
B             .018          .5            0              0
C             0             0             .5             .082
D             0             0             .082           .5

Upvotes: 1

Views: 524

Answers (1)

d.b
d.b

Reputation: 32538

#DATA
m = structure(c(50L, 1L, 0L, 0L, 1L, 6L, 0L, 0L, 0L, 0L, 51L, 8L, 
0L, 0L, 8L, 46L), .Dim = c(4L, 4L), .Dimnames = list(c("A", "B", 
"C", "D"), c("A", "B", "C", "D")))

Use sapply to go through each column and normalize

sapply(X = 1:NCOL(m), function(i) round(x = m[,i]/(m[i,i]+diag(m)), digits = 3))
#   [,1]  [,2]  [,3]  [,4]
#A 0.500 0.018 0.000 0.000
#B 0.018 0.500 0.000 0.000
#C 0.000 0.000 0.500 0.082
#D 0.000 0.000 0.082 0.500

You could substitute 1 in the diagonal elements with replace

sapply(X = 1:NCOL(m), function(i)
    replace(x = round(x = m[,i]/(m[i,i]+diag(m)), digits = 3),
            list = i,
            values = 1))
#   [,1]  [,2]  [,3]  [,4]
#A 1.000 0.018 0.000 0.000
#B 0.018 1.000 0.000 0.000
#C 0.000 0.000 1.000 0.082
#D 0.000 0.000 0.082 1.000

Upvotes: 1

Related Questions