Path: news.mathworks.com!not-for-mail
From: "hailiang shen" <hlshen2005@gmail.com>
Newsgroups: comp.soft-sys.matlab
Subject: two questions on 'gamultiobj' function
Date: Sat, 17 Oct 2009 23:49:04 +0000 (UTC)
Organization: Guelph
Lines: 6
Message-ID: <hbdl5f$lbd$1@fred.mathworks.com>
Reply-To: "hailiang shen" <hlshen2005@gmail.com>
NNTP-Posting-Host: webapp-05-blr.mathworks.com
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Trace: fred.mathworks.com 1255823344 21869 172.30.248.35 (17 Oct 2009 23:49:04 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Sat, 17 Oct 2009 23:49:04 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 1339183
Xref: news.mathworks.com comp.soft-sys.matlab:578106


Dear all, 
I have two questions on the &#8216;gamultiobj&#8217; function in obtaining Pareto front. 
1 by default, the &#8216;ParetoFraction&#8217; is 0.35. Does it mean I only take the 35% of the individuals in the first front, i.e., the one with the highest rank?
2 in calculating the crowding distance of each individual, does the algorithm used in Matlab normalize the objectives values? i.e., does it matter if one objective is of range (0,1) and the other one is (E8, 5E8)?
Thanks,
Hailiang