I'm trying to run the stemmer on the word "cleaner", and expected it to return "clean". However, it simply returns the word in its original form, "cleaner". Is this a know bug, or the desired outcome? Is there any way I can get around this? Thanks, Martin
Please see the FAQ at the end of http://tartarus.org/~martin/PorterStemmer/ At 14:05 07/03/2005 -0700, Martin May wrote: > >I'm trying to run the stemmer on the word "cleaner", and expected it to >return "clean". However, it simply returns the word in its original >form, "cleaner".
Hi all, I'm trying to get the snowball stemmers to work in Dutch on a windoze system with Activestate perl installed. I've downloaded and installed Lingua::Stem::Snowball through ppm, but only a few languages are supported (and Dutch isn't included). I've managed before to add dutch (using a modified add_stemmer.pl) and recompile Snowball on a Debian system in the blink of an eye. But since this isn't Linux (duhh) I'm unable add dutch support and recompile the snowball wrapper to work with it. Frankly, I'm quite stuck at the moment and was wondering if anyone had any hints how to proceed next. Hoping you could help
On Fri, 2005-02-25 at 15:37 +0000, Richard Boulton wrote: > Thank you for this work - it sounds helpful. I won't have a chance to > take a look at it until after the weekend, but I will make sure to do so > as soon as possible. It took a little longer than I hoped to find time to do this, but I've now applied your patches to our CVS repository. Thanks for the assistance. -- -- Richard Boulton <richard <at> tartarus.org>
Hello, > It took a little longer than I hoped to find time to do this, but I've > now applied your patches to our CVS repository. Thanks for the > assistance. You're welcome. Thanks for having applyed them. As I have mentionned in one of my previous messages, I have also an updated version of PyStemmer from http://sourceforge.net/projects/pystemmer/ if you are interested. This 'new' version of PyStemmer is supporting both Python and Jython programs. Let me know if you are interested. If yes, I will look how to integrate both interfaces (Python and Jython) in the current snowball CVS tree, as I think this will be great to have all the stuff in the same repository. Of course, I will do propositions on the mailing list before making the integration. I will also contact PyStemmer author to see what he think about this. If you are interested, and when all is ready, I will be happy to either send a patch against current CVS, or commit the changes in the CVS myself if I have an access. Good day. Olivier -- -- . __ . ___ __. | Olivier Bornet Olivier.Bornet <at> idiap.ch(Continue reading)
On Thu, 2005-03-17 at 10:15 +0100, Martin Koose wrote: > Hi, > > do you have a binary version of the german Ansi-C stemmer. > The C code does not compile properly, and the Code itself > seems to be quite complicated for non c geeks. If you're asking for a compiled version of the code, you'd need to specify what platform and architecture you are wanting to run the code on. However, we don't generally make compiled versions available, because there are so many different platforms and compilers out there that it would be impossible to support everybody anyway, and we don't have access to many of the possible platforms that people might want compiled code for. I am interested in making sure everybody can compile the code, however. If you can give a clear explanation of whatever problem it is that you are having with the compilation, I will try and be of assistance. It should be quite straightforward. -- -- Richard Boulton <richard <at> tartarus.org>
On Thu, Mar 17, 2005 at 04:42:45PM +0000, Richard Boulton wrote: > I am interested in making sure everybody can compile the code, however. > If you can give a clear explanation of whatever problem it is that you > are having with the compilation, I will try and be of assistance. It > should be quite straightforward. In snowball/snowball: ---------------------------------------------------------------------- $ make -f GNUmakefile ... gcc-3.0 -Iinclude -W -Wall -Wmissing-prototypes -Wmissing-declarations -Werror -c -o src_c/stem_english.o src_c/stem_english.c cc1: warnings being treated as errors src_c/stem_english.c: In function `r_prelude': src_c/stem_english.c:376: warning: label `lab1' defined but not used make: *** [src_c/stem_english.o] Error 1 ---------------------------------------------------------------------- However if I run the make command again, it works properly. It does, however use a different compile invocation: ---------------------------------------------------------------------- gcc-3.0 -Iinclude -O4 -c -o src_c/stem_english.o src_c/stem_english.c ---------------------------------------------------------------------- So something is weird with the makefile.(Continue reading)
On Thu, 2005-03-17 at 16:52 +0000, James Aylett wrote: > So something is weird with the makefile. Hmm - it looks like a bug in GNUmake - this happens for me with make version 3.79.1 (on ixion), but is fine with version 3.80 (on my home machine). The old make seems to be ignoring my custom rule for building the algorithms, and passing -Werror to them anyway. A workaround for now would be to remove -Werror from the makefile entirely. > (I was going to ask why you require gmake. Then I looked at the > makefile
Hello! Some time ago I sent my version of Slovene stemmer and after that small corpus of words to test the stemmer with. I guess the question goes to Martin - is there any chance you'll publish the stemmer on tartarus page? Regards, Boštjan
RSS Feed11 | |
|---|---|
13 | |
19 | |
8 | |
9 | |
1 | |
5 | |
1 | |
6 | |
8 | |
4 | |
3 | |
8 | |
2 | |
19 | |
10 | |
20 | |
7 | |
4 | |
9 | |
3 | |
5 | |
7 | |
5 | |
3 | |
11 | |
3 | |
2 | |
1 | |
3 | |
2 | |
10 | |
33 | |
14 | |
3 | |
5 | |
4 | |
15 | |
9 | |
22 | |
18 | |
7 | |
4 | |
6 | |
13 | |
15 | |
1 | |
3 | |
5 | |
13 | |
3 | |
9 | |
7 | |
9 | |
4 | |
1 | |
10 | |
5 | |
3 | |
14 | |
1 | |
7 | |
8 | |
3 | |
8 | |
5 | |
1 | |
7 | |
8 | |
12 | |
37 | |
17 | |
2 | |
35 | |
1 | |
1 | |
11 | |
10 | |
3 | |
2 | |
8 | |
9 | |
5 | |
8 | |
1 | |
5 | |
7 | |
5 | |
24 | |
11 | |
12 | |
3 | |
17 | |
20 | |
3 | |
9 | |
17 | |
3 | |
12 | |
15 | |
8 | |
24 | |
26 | |
15 | |
16 | |
31 | |
7 | |
6 | |
7 | |
32 | |
7 | |
14 | |
7 | |
147 | |
10 | |
16 | |
9 | |
1 | |
27 | |
18 | |
18 | |
32 | |
17 | |
32 | |
50 | |
5 |
