behzad@ec.ecn.purdue.edu (Behzad M Shahshahani) (04/15/91)
Has anyone out there seen any application of neural networks for estimating the number of underlying clusters in a mixture distribution? As far as know, this is a very hard statistical problem.
ntm1169@dsac.dla.mil (Mott Given) (04/15/91)
From article <1991Apr14.170640.2662@noose.ecn.purdue.edu>, by behzad@ec.ecn.purdue.edu (Behzad M Shahshahani): > > Has anyone out there seen any application of neural networks for > estimating the number of underlying clusters in a mixture distribution? I think that Nestor Inc.'s Restricted Coulomb Energy (RCE) network architecture would help to solve your problem. You can write them at Nestor Inc., One Richmond Square, Providence, Rhode Island 02906 or phone 401-331-9640. I have no connection with this company, and have not used their software. -- Mott Given @ Defense Logistics Agency Systems Automation Center, DSAC-TMP, Bldg. 27-1, P.O. Box 1605, Columbus, OH 43216-5002 INTERNET: mgiven@dsac.dla.mil UUCP: ...{osu-cis}!dsac!mgiven Phone: 614-238-9431 AUTOVON: 850-9431 FAX: 614-238-9928 I speak for myself