Completing the princeton annotated gloss corpus project

Abstract

In the Princeton WordNet Gloss Corpus, the word forms from the definitions (“glosses”) in WordNet's synsets are manually linked to the context-appropriate sense in the WordNet. The glosses then become a sense-disambiguated corpus annotated against WordNet version 3.0. The result is also called a semantic concordance, which can be seen as both a lexicon (WordNet extension) and an annotated corpus. In this work we motivate and present the initial steps to complete the annotation of all open-class words in this corpus. Finally, we introduce a freely-available annotation interface built as an Emacs extension, and evaluate a preliminary annotation effort.