Varsågod!

Here are (or will be) some resources to use in Swedish language engineering.

• Two-level description of Swedish (PCKIMMO v.1)

  Two-level description of Swedish (PCKIMMO v.2)

  The first one is a TAR file. The second one is a ZIP file. Let me know if you need them in another kind of archive.

  These are in the formats used in the first and second versions of SIL's implementation of Kimmo K.'s two-level system. The actual data is the same. The files for version two were created automatically with a tool that SIL offers.

  The mini-lexicons have names inspired by the inflectional tables in S. Hellberg. The deviations (usually more precise groupings) should be easy enough to understand.

  This is a temporary home. I will be putting it up as an Open Source project as soon as I get some time. Then we can all benefit from each other's refinements. But it's pretty good the way it is.

  It was developed for a NUTEK project ... more precise info will come.

  
  
• Brill tagger for Swedish (finally found it). This is only the datafiles. You will have to download Eric Brill's tagger from his homepage. The last time I saw it, it was on his Microsoft homepage (he's there now-a-days). This should get you there if you're having problems. Find and rename "start-state-tagger.c" to "start-state-tagger.c.orig" and use this one instead. It has been adjusted for the PAROLE tagset for Swedish.

The only thing I ask is credit where credit is due when you use the Swedish data files.