|
a |
|
b/man/plotSilhouette.Rd |
|
|
1 |
% Generated by roxygen2: do not edit by hand |
|
|
2 |
% Please edit documentation in R/DIscBIO-generic-plotSilhouette.R |
|
|
3 |
\name{plotSilhouette} |
|
|
4 |
\alias{plotSilhouette} |
|
|
5 |
\alias{plotSilhouette,DISCBIO-method} |
|
|
6 |
\title{Silhouette Plot for K-means clustering} |
|
|
7 |
\usage{ |
|
|
8 |
plotSilhouette(object, K) |
|
|
9 |
|
|
|
10 |
\S4method{plotSilhouette}{DISCBIO}(object, K) |
|
|
11 |
} |
|
|
12 |
\arguments{ |
|
|
13 |
\item{object}{\code{DISCBIO} class object.} |
|
|
14 |
|
|
|
15 |
\item{K}{A numeric value of the number of clusters} |
|
|
16 |
} |
|
|
17 |
\value{ |
|
|
18 |
A silhouette plot |
|
|
19 |
} |
|
|
20 |
\description{ |
|
|
21 |
The silhouette provides a representation of how well each point |
|
|
22 |
is represented by its cluster in comparison to the closest neighboring |
|
|
23 |
cluster. It computes for each point the difference between the average |
|
|
24 |
similarity to all points in the same cluster and to all points in the |
|
|
25 |
closest neighboring cluster. This difference it normalize such that it can |
|
|
26 |
take values between -1 and 1 with higher values reflecting better |
|
|
27 |
representation of a point by its cluster. |
|
|
28 |
} |