A semi-automatic tree annotating workbench for building a Korean treebank

Joon Ho Lim, So Young Park, Yong Jae Kwak, Hae-Chang Rim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In this paper, we propose a semi-automatic tree annotating workbench for building a Korean treebank. Generally, building a treebank requires an enormous effort by the annotator. In order to improve annotating efficiency, decrease the number of intervention required by the annotator, and help maintain consistent annotation in building a treebank, we have developed a semi-automatic tree annotating workbench consisting of following three stages: syntactic pattern extraction, syntactic pattern selection, and syntactic pattern application. The experiment was carried out with 27,966 tree tagged sentences as a training set and 3,108 sentences as a test set. As a result, the burden of manual annotation can be reduced by about 47% with the best selection of the feature set by using the proposed tree annotating workbench.

Original languageEnglish
Pages (from-to)253-257
Number of pages5
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2945
Publication statusPublished - 2004 Dec 1

    Fingerprint

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this