For two different partitioning function computes confusion matrix.
Usage
confusion.matrix(clust1, clust2)
Arguments
clust1
integer vector with information about cluster id the object is assigned to.
If vector is not integer type, it will be coerced with warning.
clust2
integer vector with information about cluster id the object is assigned to.
If vector is not integer type, it will be coerced with warning.
Details
Let P and P' be two different partitioning of the same data. Partitionings are represent as two
vectors clust1, clust2. Both vectors should have the same length.
Confusion matrix measures the size of intersection between clusters comming from P and P'
according to equation:
M[i,j] = | intersection of P(i) and P'(j) |
where:
P(i)
- cluster which belongs to partitioning P,
P'(j)
- cluster which belongs to partitioning P',
|A|
- cardinality of set A.
Value
cls.set.section returns a n x m integer matrix
where n = |P| and m = |P'| defined above.
Author(s)
Lukasz Nieweglowski
See Also
Result used in similarity.index.
Examples
# create two different subsamples
mx1 <- matrix(as.integer( c(1,2,3,4,5,6,1,1,2,2,3,3) ), 6, 2 )
mx2 <- matrix(as.integer( c(1,2,4,5,6,7,1,1,2,2,3,3) ), 6, 2 )
# find section
m = cls.set.section(mx1,mx2)
confusion.matrix(as.integer(m[,2]),as.integer(m[,3]))