Elamur Research
☰
home / papers / 1997 / GermaNet - a Lexical-Semantic Net for German

GermaNet - a Lexical-Semantic Net for German

Written for the Research library (claude-opus-5-5) from the paper text, in our own words: the license "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" does not allow redistributing the text itself. Original — aclanthology.org/W97-0802.pdf

Summary

German long lacked a freely usable, large semantic lexicon of the kind English had in WordNet. Without one, WordNet-based tools such as sense disambiguation, information extraction and retrieval could not be ported easily. The authors describe GermaNet, a wordnet for German built from scratch rather than translated from English. It keeps the Princeton WordNet database format and interface so the two remain interoperable and German can later join EuroWordNet.

The core vocabulary was selected from lemma frequency lists drawn from large German corpora and refined with CELEX. The synsets and links themselves were encoded manually with the help of dictionaries.

GermaNet departs from WordNet in several principled ways:

  • systematically marked non-lexicalized (artificial) nodes;
  • explicit cross-classification where folk and expert taxonomies conflict;
  • a pointer for regular polysemy;
  • taxonomic adjectives;
  • sense-level syntactic frames;
  • a mixed enumerative/compositional treatment of particle verbs.

At the time of writing, the net held about 16,000 words. It is meant as an ontology for semantic corpus exploration and a training resource for statistical NLP.

Key points

  • Size and coverage: about 16,000 words, aiming at the German base vocabulary. Nouns, verbs and adjectives are modeled, but adverbs are not yet. Each class is split into roughly 15 semantic fields to organize the lexicographic work.
  • Source data: a tagged, lemmatized corpus of 60 million words (the ECI corpus plus the Tübinger NewsKorpus) supplied the frequency lists. The CELEX German database was used for tuning.
  • Relation changes:
    • WordNet's entailment is split into entailment (backward presupposition) and a new subevent relation.
    • Cause may link a verb to a resultative adjective.
    • Meronymy is a single relation.
    • Selectional restrictions are added.
  • Artificial concepts and polysemy: artificial concepts are flagged with "?". A regular polysemy pointer (e.g., animal/meat) lets infrequent productive senses be inferred without separate entries.
  • Verbs:
    • Syntactic frames are derived from CELEX complementation codes and record case, reflexives and expletives.
    • Particle verbs are interpreted compositionally through a hierarchical particle lexicon, after morphological analysis with GERTWOL.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Natural language processing

Topics: Word sense disambiguation

Cited by

171 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 170 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{hamp1997germanet,
  title = {{GermaNet - a Lexical-Semantic Net for German}},
  author = {Hamp, Birgit and Feldweg, Helmut},
  booktitle = {Automatic Information Extraction and Building of Lexical Semantic Resources for NLP Applications},
  year = {1997},
  url = {https://aclanthology.org/W97-0802/}
}

References

11 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. A. Alonge. 1996. Definition of the links and subsets for verbs. Deliverable D006, WP4.1, EuroWordNet, LE2-4003. find
  2. L. Bloksma, P. Diez-Orzas, and P. Vossen. 1996. User Requirements and Functional Specification of the EuroWordNet Project. Deliverable D001, WP1, EuroWordNet, LE2-4003. find
  3. G. Burnage, 1995. The CELEX Lexical Database, Release 2. CELEX - Centre for Lexical Information; Max Planck Institute for Psycholinguistics, The Netherlands. find
  4. C. Fellbaum. 1993. A Semantic Network of English Verbs. In G.A. Miller, It. Beckwith, C. Fellbaum, D. Gross, and K. Miller, editors, Five Papers on WordNet. August. Revised version. Verb Database ,hop"°. ,..on.o, l :m:o herausl au~en (motion towards the deictic centre out o'f s.th. with manner oi motion: laui:en) l Figure 7: Particle Verbs Wolfgang Fleischer and Irmhild Barz. 1992. Wortbildung der deutsehen Gegenwartssprache. Max Niemeyer Verlag, Tiibingen. find
  5. GERTWOL. 1996. German morphological analyser. http://www.lingsoft.fi/doc/gertwol/. ↗
  6. Franz Hundsnurscher and Jochen Splett. 1982. Semantik der Adjektive des Deutschen: Analyse der semantisehen Relationen. Westdeutscher Verlag, Opladen. find
  7. European Corpus Initiative. 1994. European Corpus Initiative Multilingual Corpus. B. Levin, G. Song, and B.T.S. Atkins. in press. Making sense of corpus data: A case study of verbs of sound. International Journal o] Corpus Linguistics, page 41 pages. find
  8. G.A. Miller, R. Beckwith, C. Fellbanm, D. Gross, and K. Miller. 1993. Five Papers on WordNet. Technical report, Cognitive Science Laboratory, Princeton University, August. Revised version. find
  9. James Pustejovsky. 1996. The Generative Lexicon. MIT Press. find
  10. Rosmary Stegmann. 1996. Semantic Analysis and Classification of Verbs of Direction. MA-Thesis. Seminar ffir Sprachwissenschaft, Universit/it Tfibingen. find
  11. Barbara Stiebels. 1994. Lexikalische Argumente und Adjunkte. Zum semantischen Beitrag von Particle Database find