Hi Charles, You're solution is great (and is actually what my professor suggested me to do today).
In the meantime I searched even more and found this article: http://davis.wpi.edu/~xmdv/docs/tr0313_osf.pdf That gives a good description of the problem and of his attempts at solving it. I started by implementing (in a very rough way) the solution of "all possible combinations", but it is not relevant for more then 7 dimensions. His solution will involve a mix of your solution with a local optimum, but I don't think I will go more into solving it anytime soon (but instead, make due with your solution). Thanks again, Tal ----------------Contact Details:------------------------------------------------------- Contact me: [email protected] | 972-52-7275845 Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | www.r-statistics.com/ (English) ---------------------------------------------------------------------------------------------- On Sun, Jan 3, 2010 at 10:05 PM, Charles C. Berry <[email protected]>wrote: > On Sun, 3 Jan 2010, Tal Galili wrote: > > Hi Charles, >> Thanks for answering - you are right about the posting guide (sorry). >> >> In any case, my question wasn't on how to reorder the columns (that is >> simple), but on how to choose what order to put them in. >> > > > OK, and I see you included a reference, so there is probably more to this > than meets my eye. > > But would something as simple as this be good enough?? > > library(MASS) >> hc1 <- hclust(dist(cor(log(iris[, 1:4])))) >> parcoord(log(ir)[, hc1$order ], col = 1 + (0:149)%/%50) >> > > or possibly > > hc1 <- hclust(dist( abs( cor(log(iris[, 1:4]))))) > > > Chuck > > >> >> Thanks, >> Tal >> >> >> >> ----------------Contact >> Details:------------------------------------------------------- >> Contact me: [email protected] | 972-52-7275845 >> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | >> www.r-statistics.com/ (English) >> >> ---------------------------------------------------------------------------------------------- >> >> >> >> >> On Sun, Jan 3, 2010 at 7:05 PM, Charles C. Berry <[email protected] >> >wrote: >> >> On Sat, 2 Jan 2010, Tal Galili wrote: >>> >>> Hello all, >>> >>>> >>>> I am searching for a way in R to re-order variables before presenting >>>> them >>>> in a parallel coordinates plot. >>>> >>>> So far I didn't find anything within a R related context on how to do >>>> this. >>>> I did find some texts talking about how it should be done in general, >>>> here >>>> is such example: >>>> http://tinyurl.com/ycnsjpe >>>> >>>> Is there a package or an example of the variable ordering (for parallel >>>> coordinate plot) in R ? >>>> >>>> >>> Follow the posting guide >>> >>> ??parallel >>> library(MASS) >>> ?parcoord >>> example( parcoord ) >>> >>> See the last line of the example. >>> >>> HTH, >>> >>> Chuck >>> >>> >>> Thanks, >>>> Tal >>>> >>>> >>>> ----------------Contact >>>> Details:------------------------------------------------------- >>>> Contact me: [email protected] | 972-52-7275845 >>>> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) >>>> | >>>> www.r-statistics.com/ (English) >>>> >>>> >>>> ---------------------------------------------------------------------------------------------- >>>> >>>> [[alternative HTML version deleted]] >>>> >>>> >>>> ______________________________________________ >>>> [email protected] mailing list >>>> https://stat.ethz.ch/mailman/listinfo/r-help >>>> PLEASE do read the posting guide >>>> http://www.R-project.org/posting-guide.html >>>> and provide commented, minimal, self-contained, reproducible code. >>>> >>>> >>>> Charles C. Berry (858) 534-2098 >>> Dept of Family/Preventive >>> Medicine >>> E mailto:[email protected] UC San Diego >>> http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego >>> 92093-0901 >>> >>> >>> >>> >> [[alternative HTML version deleted]] >> >> ______________________________________________ >> [email protected] mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> >> > Charles C. Berry (858) 534-2098 > Dept of Family/Preventive > Medicine > E mailto:[email protected] UC San Diego > http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego 92093-0901 > > > [[alternative HTML version deleted]] ______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
