similarity: Similarity Matrices

Description Usage Arguments Value Author(s) References Examples

Description

cts computes the connected triple based similarity matrix, srs computes the simrank based similarity matrix, and asrs computes the approximated simrank based similarity matrix.

Usage

1
2
3
4
5
cts(E, dc)

srs(E, dc, R)

asrs(E, dc)

Arguments

E

an N by M matrix of cluster ensembles

dc

decay factor, ranges from 0 to 1 inclusive

R

number of iterations for srs

Value

an N by N CTS, SRS, or ASRS matrix

Author(s)

Johnson Liu, Derek Chiu

References

MATLAB functions cts, srs, asrs in package LinkCluE by Simon Garrett

Examples

1
2
3
4
5
6
set.seed(1)
E <- matrix(rep(sample(1:4, 800, replace = TRUE)), nrow = 100)
CTS <- cts(E = E, dc = 0.8)
SRS <- srs(E = E, dc = 0.8, R = 3)
ASRS <- asrs(E = E, dc = 0.8)
purrr::walk(list(CTS, SRS, ASRS), str)

Example output

 num [1:100, 1:100] 1 0.848 0.767 0.794 0.78 ...
 num [1:100, 1:100] 1 0.0558 0.0307 0.0309 0.0307 ...
 num [1:100, 1:100] 1 0.66 0.609 0.646 0.646 ...

diceR documentation built on June 11, 2018, 5:04 p.m.