Hi José

my co-authors and I have used K-valued FCA (where K is an idempotent semifield) extensively for confusion matrix analysis (for multiclass classifiers). The technique actually applies to contingency matrices so it may apply to some ecological 2-mode data. You can find our most elaborate effort in the Journal of the Acoustics Society of America (check my publications).

We have also done some work (published in ICFCA) on gene-expression data analysis with K-FCA. The original inspiration for this was presented by Motameny et al. in ICFCA 2008 (at Montreal, I believe). You can find example of analysis in the ICFCA and CLA conference series (and in our publications in ResearchGate, of course).

And a Ph.D. student of ours has an open tool for this where he has uploaded some GE data in the link below.

Please do not hesitate to contact us if you need more information/examples.

Regards,

http://www.webgenekfca.com/