The SPECIALIST Lexicon

Antonym Generation for LEX Model

shell>cd ${ANTONYM_DIR}/bin
shell>GetAntonyms ${YEAR}

LEX model: LexRecord with Negations
Use the latest Lexicon

OptionDescriptioninputOutputNotesOption
10
  • generate antonym candidates from the Lexicon (with negative tags)
  • Lexicon.GenAntCandFromLexicon.java
  • ${LEX_DIR}/input/LEXICON
  • ${ANT_DIR}/input/antCand.data.tag
  • ${ANT_DIR}/input/domain.data
  • ./output/antCandLexicon.data
  • ./output/Cand/antCandLexicon.data.tag
  • ./output/Cand/antCandLexicon.data.tbd
  • ./output/candTagged/antCandLexicon.data.tag.tagged
  • mkdir needed directories (Cand, candTagged)
  • copy needed files
  • send antCandLexicon.data.tbd to linguist to tag, add to antCand.data.tag, and re-run until tbd = 0
  • copy antCandLexicon.data.tag.tagged to antCandLexicon.data.tag.tagged.${YEAR}
    This is antonyms from the source of LEX
10
11
  • Validate and fix tags of antonym candidates (LEX)
  • Antonym.ValidateTaggedCand.java
  • ./output/candTagged/antCandLexicon.data.tag.tagged
  • ${ANT_DIR}/input/domain.data
  • ./output/candTagged/antCandLexicon.data.tag.fixed
  • Make sure the tag and fixed files are the same
11
12
  • update release antonyms tagged file from LEX
  • automatically assign type to [NA] and domain to [DOMAIN_NONE] if Canon is [N]
  • check for new domains
  • Antonym.UpdateAllTaggedFile.java
  • ./output/candTagged/antCandLexicon.data.tag.tagged.${YEAR}
  • ${ANT_DIR}/input/antCand.data.tag.${YEAR}
  • ${ANT_DIR}/input/domain.data
  • ${ANT_DIR}/input/antCand.data.tag.updated
  • The step auto-update all antonym candidate tag file
  • The output file is used to generate antonym and negation files for the release.
  • Manully copy antCand.data.tag.updated to antCand.data.tag.updated.LEX
    cp -rp ${ANT_DIR}/input/antCand.data.tag.updated ${ANT_DIR}/input/antCand.data.tag.updated.1.LEX
12