mSigTools (version 1.0.7)

match_two_sig_sets: Find an optimal matching between two sets of signatures subject to a maximum distance.

Description

Find an optimal matching between two sets of signatures subject to a maximum distance.

Usage

match_two_sig_sets(
  x1,
  x2,
  method = "cosine",
  convert.sim.to.dist = function(x) {
     return(1 - x)
 },
  cutoff = 0.9
)

Value

A list with the elements

• table Table of extracted signatures that matched a reference signature. Each row contains the extracted signature name, the reference signature name, and the distance of the match.

• orig.matrix The matrix of numeric distances between x1 and x2.

• modified.matrix The argument orig.matrix with distances > cutoff changed to very large values.

Arguments

x1

A numerical-matrix-like object with columns as signatures.

x2

A numerical-matrix-like object with columns as signatures. Needs to have the same number of rows as x1.

method

As for the distance function in package philenropy.

convert.sim.to.dist

If method specifies a similarity rather than a distance, use this function to convert the similarity to a distance.

cutoff

A maximum distance or minimum similarity over which to pair signatures between x1 and x2.

Details

Match signatures between x1 and x2 using the function solve_LSAP, which uses the "Hungarian" (a.k.a "Kuhn–Munkres") algorithm https://en.wikipedia.org/wiki/Hungarian_algorithm, which optimizes the total cost associated with the links between nodes. This function generates a distance matrix between the two sets of signatures using method and, if necessary, convert.sim.to.dist. It then sets distances > cutoff to very large values and then applies solve_LSAP to the resulting matrix to compute a matching between x1 and x2 that minimizes the sum of the distances.

Examples

ex.sigs <- matrix(c(0.2, 0.8, 0.3, 0.7, 0.6, 0.4), nrow = 2)
colnames(ex.sigs) <- c("ex1", "ex2", "ex3")
ref.sigs <- matrix(c(0.21, 0.79, 0.19, 0.81), nrow = 2)
colnames(ref.sigs) <- c("ref1", "ref2")
match_two_sig_sets(ex.sigs, ref.sigs, cutoff = .9)