I have a big dataset which has a lot of zeros. I want to use a fuzzy clustering algorithm. Do you know a useful fuzzy clustering for sparse dataset?
You could use FCM(Fuzzy C-Means) as a good candidate.
You could use Matlab/Octave/R for implementation, they have good libraries/packages for FCM.
You don't mention if the variables are continuous or discrete or how the zeros arise. If the variables are discrete then Latent Class Analysis may be appropriate. See http://www.john-uebersax.com/stat/
Fuzzy Clustering Level Analysis via Statistical Scheme Applying Akaike’s Information Criterion, Shuya Kanagawa and Kimiaki Shinkai, International Journal of Innovative Computing, Information and Control, Vol.4, (2008), pp.2523-2531.
Please see the above paper. We tried to find reasonable clustering level using AIC.
Fuzzy G-G clustering that considers both changes in shape and volume of the clusters
If You can access also to italian literature, it's rather useful a 1998 contribute of our National Research Council, IRRPS Institute.
See the link (http://www.irpps.cnr.it/it/system/files/wp2_98_0.pdf).
Best Regards
- Badges
- Science topic