10.01.2015 Views

Package 'fpc' - open source solution for an Internet free, intelligent ...

Package 'fpc' - open source solution for an Internet free, intelligent ...

Package 'fpc' - open source solution for an Internet free, intelligent ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

48 distcritmulti<br />

distcritmulti<br />

Dist<strong>an</strong>ce based validity criteria <strong>for</strong> large data sets<br />

Description<br />

Usage<br />

Approximates average silhouette width or the Pearson version of Hubert’s gamma criterion by hacking<br />

the dataset into pieces <strong>an</strong>d averaging the subset-wise values, see Hennig <strong>an</strong>d Liao (2010).<br />

distcritmulti(x,clustering,part=NULL,ns=10,criterion="asw",<br />

fun="dist",metric="euclide<strong>an</strong>",<br />

count=FALSE,seed=NULL,...)<br />

Arguments<br />

x<br />

clustering<br />

part<br />

ns<br />

Value<br />

criterion<br />

fun<br />

metric<br />

count<br />

seed<br />

cases times variables data matrix.<br />

vector of integers indicating the clustering.<br />

vector of integer subset sizes; sum should be smaller or equal to the number of<br />

cases of x. If NULL, subset sizes are chosen approximately equal.<br />

integer. Number of subsets, only used if part==NULL.<br />

"asw" or "pearsongamma", specifies whether the average silhouette width or<br />

the Pearson version of Hubert’s gamma is computed.<br />

"dist" or "daisy", specifies which function is used <strong>for</strong> computing dissimilarities.<br />

passed on to dist (as argument method) or daisy to determine which dissimilarity<br />

is used.<br />

logical. if TRUE, the subset number just processed is printed.<br />

integer, r<strong>an</strong>dom seed. (If NULL, result depends on r<strong>an</strong>dom numbers.)<br />

... further arguments to be passed on to dist or daisy.<br />

A list with components crit.overall,crit.sub,crit.sd,part.<br />

crit.overall<br />

crit.sub<br />

crit.sd<br />

subsets<br />

value of criterion.<br />

vector of subset-wise criterion values.<br />

st<strong>an</strong>dard deviation of crit.sub, c<strong>an</strong> be used to assess stability.<br />

list of case indexes in subsets.<br />

Author(s)<br />

Christi<strong>an</strong> Hennig http://www.homepages.ucl.ac.uk/~ucakche

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!