| Type: | Package |
| Title: | Check the Clustering Tendency |
| Version: | 1.7 |
| Description: | Calculate some statistics aiming to help analyzing the clustering tendency of given data. In the first version, Hopkins statistic is implemented. See Hopkins and Skellam (1954) <doi:10.1093/oxfordjournals.aob.a083391>. |
| License: | MIT + file LICENSE |
| NeedsCompilation: | no |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.2.3 |
| Packaged: | 2023-08-19 14:18:10 UTC; wrightkevi |
| Author: | Kevin Wright 
[aut, cre],
Luo YiLan [aut],
Zeng RuTong [aut] |
| Maintainer: | Kevin Wright <kw.stat@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2023-08-19 14:42:39 UTC |
Calculate Hopkins statistic.
Description
Calculate Hopkins statistic of given data.
Usage
hopkins(X, n = ceiling(nrow(X)/10))
Arguments
X |
Data (matrix or data.frame) to check clusterability. |
n |
The number of rows to sample from X. The default is 1/10th the number of rows of X. |
Details
Note: Package clustertend is deprecated.
Use package hopkins instead.
Sample data must be preprocessed into dataframe or matrix form before given as the value of parameter "data".
Value
The value returned is actually 1-Hopkins statistic.
Author(s)
Kevin Wright, Luo YiLan, Zeng RuTong.
References
Lawson, R.G. and Jurs, P.C.(1990). New index for clustering tendency and its application to chemical problems. Journal of Chemical Information and Computer Sciences. 30(1):36-41.
Examples
set.seed(1)
hopkins(iris[,-5], n=15)