Re: [R] define number of clusters in kmeans/apcluster analysis

2015-12-15 Thread Ulrich Bodenhofer
Dear Luigi, As the others have replied already, you cannot expect a clustering algorithm to produce exactly the result that you expect intuitively. The results of clustering algorithms depend largely on the parameters and, even more importantly, on the distance/similarity measure that is used.

Re: [R] Probing a protein sequence alignment in R

2015-11-25 Thread Ulrich Bodenhofer
Hi Debra, As already noted by Boris, the right packages can be found in Bioconductor, namely Biostrings (for handling sets of sequences and pairwise alignments) and msa (for multiple alignments; a package I am maintaining). Your question does not yet clearly indicate to me whether pairwise or

[R] Version 1.4.0 of the 'apcluster' package

2014-12-03 Thread Ulrich Bodenhofer
y matrices. For more details, see the package documentation and the following URLs: http://www.bioinf.jku.at/software/apcluster/ http://cran.r-project.org/web/packages/apcluster/index.html Best regards, Ulrich Bodenhofer *

[R] Version 1.3.5 of apcluster package

2014-06-29 Thread Ulrich Bodenhofer
/apcluster/index.html Best regards, Ulrich Bodenhofer *Dr. Ulrich Bodenhofer* Associate Professor Institute of Bioinformatics *Johannes Kepler University* Altenberger Str. 69 4040 Linz, Austria Tel. +43 732 2468 4526 Fax +43 732

[R] Version 1.3.3 of apcluster package (+ Rcpp compatibility issue)

2014-02-24 Thread Ulrich Bodenhofer
*Dr. Ulrich Bodenhofer* Associate Professor Institute of Bioinformatics *Johannes Kepler University* Altenberger Str. 69 4040 Linz, Austria Tel. +43 732 2468 4526 Fax +43 732 2468 4539 bodenho...@bioinf.jku.at <mailto:bodenho...@bioinf.jku.at>

[R] Version 1.3.2 of apcluster package (+ last announcement of Webinar)

2013-06-12 Thread Ulrich Bodenhofer
DiGiacomo, Jr. Best regards, Ulrich *Dr. Ulrich Bodenhofer* Associate Professor Institute of Bioinformatics *Johannes Kepler University* Altenberger Str. 69 4040 Linz, Austria Tel. +43 732 2468 4526 Fax +43 732 2468

[R] apcluster webinar: Thursday, June 13, 2013, 7:00pm CEST

2013-05-14 Thread Ulrich Bodenhofer
Bodenhofer *Dr. Ulrich Bodenhofer* Associate Professor Institute of Bioinformatics *Johannes Kepler University* Altenberger Str. 69 4040 Linz, Austria Tel. +43 732 2468 4526 Fax +43 732 2468 4539 bodenho...@bioinf.jku.at http

[R] [R-pkgs] Version 1.3.1 of apcluster package on CRAN

2013-04-27 Thread Ulrich Bodenhofer
://www.bioinf.jku.at/software/apcluster/ http://cran.r-project.org/web/packages/apcluster/index.html Best regards, Ulrich *Dr. Ulrich Bodenhofer* Associate Professor Institute of Bioinformatics *Johannes Kepler University* Altenberger Str

Re: [R] Accessing examplars in apcluster (apcluster package)

2013-04-04 Thread Ulrich Bodenhofer
Dear Sachin, If you need the exemplars for further processing, you can access them via the 'exemplars' slot of the resulting APResult object. See the following example: > apres <- apcluster(negDistMat(r=2), x, details=TRUE) > apres APResult object Number of samples = 150 Number of iter

[R] R broken after upgrade to 2.15.3 (Ubuntu 12.10, 64bit)

2013-03-04 Thread Ulrich Bodenhofer
Hi, I am using R on Ubuntu 12.10 (64bit). This morning, Ubuntu's software updater automatically installed updates to R's base system (version 2.15.3; via the CRAN PPA). Now R does not work anymore. Here is what I get when I simply enter "R" on the shell prompt: bodenhof FUKUOKA~>R cannot fin

[R] [R-pkgs] Version 1.3.0 of apcluster package on CRAN

2013-01-09 Thread Ulrich Bodenhofer
are/apcluster/> http://cran.r-project.org/web/packages/apcluster/index.html <http://cran.r-project.org/web/packages/apcluster/index.html> Best regards, Ulrich ---- *Dr. Ulrich Bodenhofer* Associate Professor Institute of Bi

Re: [R] SVM. How to use categorical attributes?

2012-03-28 Thread Ulrich Bodenhofer
Sorry, I forgot to mention the following: all I wrote is only valid as long as your number of samples is smaller than the number of different words. If the number of samples exceeds the total number of different words, you should better use the explicit matrix representation and use some kernel (e.

Re: [R] SVM. How to use categorical attributes?

2012-03-28 Thread Ulrich Bodenhofer
Alex, To avoid the memory issue, you can directly use a "bag of words" kernel (which corresponds to using the linear kernel on the sparse bag of words matrix Steve suggested). Just a little toy example how this is done for two : > x1 <- c("how", "to", "grow", "tree") > x2 <- c("where", "to", "go

Re: [R] cluster analysis and supervised classification: an alternative to knn1?

2010-05-27 Thread Ulrich Bodenhofer
> > What do you suggest in order to assign a new observation to a determined > cluster? > As I mentioned already, I would simply assign the new observation to the cluster to whose exemplar the new observation is most similar to (in a knn1-like fashion). To compute these similarities, you can use t

Re: [R] cluster analysis and supervised classification: an alternative to knn1?

2010-05-27 Thread Ulrich Bodenhofer
Sorry, Joris, I overlooked that you already mentioned daisy() in your posting. I should have credited your recommendation in my previous message. Cheers, Ulrich -- View this message in context: http://r.789695.n4.nabble.com/cluster-analysis-and-supervised-classification-an-alternative-to-knn1-t

Re: [R] cluster analysis and supervised classification: an alternative to knn1?

2010-05-27 Thread Ulrich Bodenhofer
> > I had a look at the documentation of the package apcluster. > That's interesting but do you have any example using it with both > categorical > and numerical variables? I'd like to test it with a large dataset.. > Your posting has opened my eyes: problems where both numerical and categorical f

Re: [R] cluster analysis and supervised classification: an alternative to knn1?

2010-05-27 Thread Ulrich Bodenhofer
abanero wrote: > > Do you know something like “knn1” that works with categorical variables > too? > Do you have any suggestion? > There are surely plenty of clustering algorithms around that do not require a vector space structure on the inputs (like KNN does). I think agglomerative clustering w