Standardizing Wordnets in the ISO Standard LMF: Wordnet-LMF for GermaNet

Proceedings of the 23rd International Conference on Computational Linguistics (Coling 2010), pages 456–464, Beijing, August 2010 Standardizing Wordnets in the ISO Standard LMF: Wordnet-LMF for GermaNet Verena Henrich Universtiy of Tübingen Department of Linguistics verena.henrich@uni- tuebingen.de Erhard Hinrichs Universtiy of Tübingen Department of Linguistics erhard.hinrichs@uni- tuebingen.de Abstract It has been recognized for quite some time that sustainable data formats play an important role in the development and curation of linguistic resources. The purpose of this paper is to show how GermaNet, the German version of the Princeton WordNet, can be con- verted to the Lexical Markup Frame- work (LMF), a published ISO standard (ISO-24613) for encoding lexical re- sources. The conversion builds on Wordnet-LMF, which has been pro- posed in the context of the EU KYOTO project as an LMF format for wordnets. The present paper proposes a number of crucial modifications and a set of extensions to Wordnet-LMF that are needed for conversion of wordnets in general and for conversion of Ger- maNet in particular. 1 Introduction It has been recognized for quite some time that sustainable data formats play an important role in the development and curation of linguistic resources. As witnessed by the success of the guidelines of the Text Encoding Initiative 1 (TEI) and of published standards issued by the International Standards Organization 2 (ISO), markup languages such as XML3 (short for: Extensible Markup Language) have become lingua francas for encoding linguistic resources of different types, including phonetic transcrip- 1 See http://www.tei-c.org 2 See http://www.iso.org 3 See http://www.w3.org/TR/REC-xml/ tions, (annotated) text corpora, and dictionar- ies. It is fair to say that it has become common practice among developers of new linguistic resources to consult TEI guidelines and ISO standards in order to develop standard- conformant encoding schemes that serve as an interchange format and that can be docu- mented and validated by Document Type Definitions (DTD) and XML schemata. However, for resources that were developed prior to or largely in parallel with the emerging acceptance of markup languages and of emerg- ing encoding standards, the situation is far more heterogeneous. A wide variety of legacy formats exists, many of which have persisted due to existing user communities and the availability of tools that can process only such idiosyncratic formats. The development of wordnets for a large number of languages is a typical example of a type of linguistic re- source, where legacy formats still persist as a de facto standard. WordNet 1.6 is encoded in the data format of lexicographer files4 that was designed for the English Princeton WordNet (Fellbaum, 1998). It is a plain-text format for storing wordnet data and allows lexicographers to encode lexical and conceptual relations among lexical units and synsets by use of spe- cial-purpose diacritics. There exist numerous tools that can process WordNet 1.6 lexicogra- pher files to extract relevant information or to transform the data into other special-purpose formats such as Prolog-fact databases. Even tough still widely used for the reasons just mentioned, the complexity of the format itself has …

Content truncated to 500 words