search for: 4000x3

Hi, I'm trying to use R to cluster words with related meanings. Does anyone know of a non-hierarchical clustering method in R that produces non-exclusive clusters? With non-exclusive, I mean that words should be allowed to be part of multiple clusters. So my data matrix would look something like: T1 T2 T3 CLOWN_N 0 1 0 BANK_N 3 0 2 RIVER_N 0 0 2 FLOW_V 0 0 3 MONEY_N 2 0 0 PAY_V 2 0 0 The

...) outputs probabilities > of membership in each cluster. > > > the one above, its dimensions would be in the order of (100000, > > 100000). Does anyone know if this would cause practical problems, > > perhaps very slow clustering? > > I had a much smaller matrix, 4000x3, fanny took about 4 > minutes wall clock time on a lightly loaded (there were many > other processes, but none > computational) 1.4 GHz Athlon, It was completely CPU-bound. > > -- > bhaskar > > ______________________________________________ > R-help at stat.math.ethz...

...but practically infeasible in S, unless > they are coded in C or Fortran and called from S. These problems seem to run > fine in R. Some but not all. Last night I was trying to run my survival examples in R on Windows, including a large loop or alternatively a vectorized version using many 4000x3 matrices. Both run happily in S-PLUS 4.5 in about three minutes and less than 10Mb allocated. R needs a largish heap (20Mb failed, 30Mb seemed enough) that got partially swapped out (the machine has 32Mb RAM) and I gave up after an hour of thrashing on the loop version, and the vectorized version...