Improvements on ELCmapas and Complementa tools

Good news for the users of some of the most popular CAPFITOGEN tools!.

ELCmapas tool is now able to perform four new methods to determine the optimal number of clusters:

kmeansbic: A simple system that performs successive K – means clustering procedures (increasing the number of clusters) and determining for each the Bayesian Information Criterion (BIC).

elbow: K – means as a clustering algorithm where the cut-off point is determined on the basis of the decrease in the sum of the intra-group squares.

medoides: Method of partition clustering around the medoids (pam).

calinski: Calinski-Harabasz criterion to select the optimal number of clusters from applying Kmeans method in “cascade” ans specifying the number of iterations.

ssi: “Simple Structure Index” criterion to select the optimal number of clusters from applying Kmeans method in “cascade” ans specifying the number of iterations.

bic: Optimal number of clusters is established for parameterized Gaussian mixture models initialized by model-based hierarchical clustering using “Bayesian information criterion”.

About Complementa tool, now is able to perform complementarity and coverage analysis extracting in the same process the ELC categories from an ELC map (the previous use of Representa tool is avoided), and can exclude spatial duplicates (two close presence sites within the same taxon dataset which would represent the same population). Therefore Complementa becomes an independent tool.

These improvements are available in the latest update of the tools, please brush up our page for download it.