Representation of Polarity Information of Elements of German Compound Words

Thierry Declerck

In: John P. McCrae , Christian Chiarcos , Elena Montiel Ponsoda , Thierry Declerck , Petya Osenova , Sebastian Hellmann (Hrsg.). Proceedinga LDL 2016 5th Workshop on Linked Data in Linguistics: Managing, Building and Using Linked Language Resources. Workshop on Linked Data in Linguistics (LDL-16) befindet sich Tenth International Conference on Language Resources and Evaluation (LREC'16) May 24-24 Portoroz Slovenia ELRA 5/2016.


We present on-going work on using formal representation frameworks for encoding polarity information that can be attached to elements of German compound words. As a departure point we have a polarity lexicon for German words that was compiled and ranked on the basis of the integration of four pre-existing polarity lexicons that were available in different formats. As for the formal representation frameworks we are considering for the encoding of the lexical data the lexicon model for ontologies (lemon), more specifically its modules ontolex (Ontology-lexicon interface) and decomp (Decomposition), which have been developed in the context of the W3C Ontology-Lexica Community Group. For the encoding of the polarity information we adopt a slightly modified version of the Marl ontological modelling, developed at the Universidad Politécnica de Madrid.


LDL2016-Compounds-Polarity-Final.pdf (pdf, 452 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence