Skip to content
FrancisBond edited this page Oct 10, 2011 · 11 revisions

This is a page for up-to-date notes on running the Korean Resource Grammar.

The Korean Resource Grammar was originally built by JongBokKim and JaehyungYang. A revised matrix-compliant version has been rebuilt by SanghounSong and FrancisBond --- the KRG now both parses and generates.

Encoding Issues

  • CLIM does not currently support entering Korean directly under linux. Instead, we recommend you enter it though emacs.
  • To show Korean in the output either
    • use pangolui (see Alternative Lui Implementations on the LkbLui page) (FrancisBond recommends this)

    • specify Korean fonts in the .luirc (see LuiRc)

Tokenizing

  • There are three script files; lkb/script, lkb/script.common, lkb/test.
  • lkb/script: If you take this, the input string should be fully tokenized.
  • lkb/script.common: If you take this, you can input just common Korean sentences. This script will operate the pre-processor for parsing and generation.
  • lkb/test: If you want to test each grammar module with a small size of lexicon, please take this.