Model Based Diagnostics for Multivariate Cluster Analysis

Assessment and diagnostics for comparing competing clustering solutions, using predictive models. The main intended use is for comparing clustering/classification solutions of ecological data (e.g. presence/absence, counts, ordinal scores) to 1) find an optimal partitioning solution, 2) identify characteristic species and 3) refine a classification by merging clusters that increase predictive performance. However, in a more general sense, this package can do the above for any set of clustering solutions for i observations of j variables.



Build StatusCode coverage

An R package for assessment and diagnostics of competing clustering solutions, using predictive models. The main intended use is for comparing clustering/classification solutions of ecological data (e.g. presence/absence, counts, ordinal scores) to:

  1. find an optimal partitioning solution
  2. identify characteristic species and
  3. refine a classification by merging clusters such that it increases predictive performance.

However, in a more general sense, this package can do the above for any set of clustering solutions for i observations of j variables. More details on the background and theory behind using predictive models for classification assessment, in an ecological context, can be found in Lyons et al. (2016).

Installation

In R, simply use

install.packages("optimus")

See the package page on CRAN for more details:
https://cran.r-project.org/package=optimus

Development version

If you want to install the development version of optimus, for example if I've added something new that you want to use, but it's not yet up on CRAN, then you can also install directly from github. It's very easy - simply use Hadley Wickham's (excellent) devtools package - install devtools from CRAN within R using

install.packages("devtools")

then call

library(devtools)
devtools::install_github("mitchest/optimus")

Bugs

There are some probably. If you find them, please let me know about them - either directly on github, or the contact details below.

How to use optimus?

You can find the vignette on the CRAN home page, or you can access it here too (might be new things here before CRAN occasionally).
Check out the tutorial here.

Contact

References

Lyons et al. 2016. Model-based assessment of ecological community classifications. Journal of Vegetation Science: 27 (4) 704--715. DOI: http://dx.doi.org/10.1111/jvs.12400

News

optimus 0.2.0

  • Nothing users should notice in functionality
  • Couple new tests based on feedback
  • Users might notice better speeds for some tasks
  • Check for changes in between CRAN releases: https://github.com/mitchest/optimus

optimus 0.1.0

  • Added a NEWS.md file to track changes to the package.
  • This is the first release on CRAN - the package is now at the point where I am not making changes very frequently
  • This version now has a vignette, so that should be your first port of call
  • There has been no major functionality change since Sept 2016, just bug fixes and optimisations
  • Check for changes in between CRAN releases: https://github.com/mitchest/optimus

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("optimus")

0.2.0 by Mitchell Lyons, 2 years ago


https://github.com/mitchest/optimus/


Report a bug at https://github.com/mitchest/optimus/issues


Browse source code at https://github.com/cran/optimus


Authors: Mitchell Lyons [aut, cre]


Documentation:   PDF Manual  


GPL-3 license


Imports stats, methods, mvabund, ordinal

Suggests testthat, knitr, rmarkdown


See at CRAN