Switch branches/tags
Nothing to show
Commits on Jul 6, 2012
  1. Include makefiles via $KORP_INSTALL to allow running outside $KORP_IN…

    Also add dependency relation fields.
    committed Jul 6, 2012
Commits on Jun 26, 2012
  1. Add subcategorisation frames relation extraction using syntactic depe…

    Delete, the first experiment with term extraction and lemmatisation.
    committed Jun 26, 2012
Commits on Jun 11, 2012
  1. Add rough program for extracting subclass relation annotations.

    Also update to take command line arguments.
    committed Jun 11, 2012
Commits on Jun 8, 2012
  1. Move the term candidate head annotation jape rule to its own file.

    Update term candidate rule to work on the Korp annotations.
    committed Jun 8, 2012
  2. Add command line tool to make and populate a very big GATE corpus.

    GATE Developer opens each file as it populates the corpus, then runs
    out of RAM. I could probably have upped GATE's memory too, but this
    is handy anyway.
    committed Jun 8, 2012
Commits on Jun 7, 2012
Commits on Jun 1, 2012
  1. Add "Head" annotation to last noun in TermCandidate.

    Use appelt matching type to match the single longest match possible
    whenever matches can overlap.
    committed Jun 1, 2012
Commits on May 30, 2012
Commits on Apr 3, 2012
  1. Decently working lemma fetching and choosing from Saldo

    Basically gets an analysis list from Saldo, then chooses the first lemma with
    the same POS and morphology for all common nouns.
    Adds a module for parsing SUC, Parole and Saldo POS+morphology tags.
    Removes first attempt at a Saldo library.
    New saldo Library at
    Looks like 687 lookups takes 20 minutes and has 24 unknowns. Sweet!
    committed Apr 3, 2012
Commits on Mar 27, 2012
Commits on Mar 23, 2012
  1. Add Olt - first exploration of GATE data and term extraction using TF

    frequency done on term on all nouns and pronouns.
    next step - lemmatize and do TF on lemmas.
    committed Mar 23, 2012
Commits on Mar 22, 2012
  1. Working local setup and some longer term changes to Tagger_Framework

    Change TaggerFramework to Tagger_Framework
    Add option to GenericTagger for inserting blank line between sentences.
    Right now this is implemented as newline after the string annotation is a .
    Add a working local configuration and a fairly sane version of hunpostagger.gapp
    and clear the cruft from hunpostagger.gapp itself.
    Remove the TF jar, ant can make it easily.
    committed Mar 22, 2012
  2. Some changes for Hunpos

    Quite specific to my setup.
    committed Mar 22, 2012
Commits on Mar 21, 2012
  1. Initial commit.

    Add some READMEs
    Add unchanged Tagger_Framework from
    committed Mar 21, 2012