home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!news.bbn.com!news.bbn.com!aboulang
- From: aboulang@bbn.com (Albert Boulanger)
- Newsgroups: comp.ai.neural-nets
- Subject: Re: clustering, genetic algorithm
- Message-ID: <ABOULANG.92Aug22134345@wilma.bbn.com>
- Date: 22 Aug 92 20:43:45 GMT
- References: <1992Aug11.185653.2652@noose.ecn.purdue.edu>
- Reply-To: aboulanger@bbn.com
- Organization: BBN, Cambridge MA
- Lines: 24
- NNTP-Posting-Host: wilma.bbn.com
- In-reply-to: whsu@rhino.ecn.purdue.edu's message of Tue, 11 Aug 1992 18:56:53 GMT
-
- In article <1992Aug11.185653.2652@noose.ecn.purdue.edu> whsu@rhino.ecn.purdue.edu
- (William Hsu) writes:
-
-
- I am interested to know of any reference of using genetic type algorithm to
- do clustering. Thanks
- --
- William Hsu Department of Electrical Engineering, Purdue University
- West Lafayette, IN 47907. Tel : (317) 494-3471.
- Internet: whsu@panther.ecn.purdue.edu
- uucp: ...!{decvax,hplabs,ihnp4,linus,rutgers,seismo}!pur-ee!whsu
-
-
- Jim Kelly at BBN and Dave Davis at TICA played with a "Weighted"
- k-nearest neighbors idea. (Stricly speaking this was a supervised
- method.) The weights were optimised by a GA. Here is a reference:
-
-
-
- "Using Machine Learning Techniques to Visualize and Refine Criteria
- for Biological Integrity" (with Kenneth Anderson, Lawrence Davis,
- Herbert Gish, James Kelly, & Jeffrey Morrill). In Proceedings of
- Biological Criteria: Research and Regulation Symposium, US EPA, Office
- of Water, Washington DC, July 1991, 123-128.
-