Adding Polarity Information to Entries of the Database of Bavarian Dialects in Austria

Thierry Declerck; Amelie Dorn; Eveline Wandl-Vogt

In: Tinatin Margalitadze; George Meladze (Hrsg.). Proceedings of the XVII EURALEX International Congress. EURALEX International Congress (EURALEX-16), Lexicography and Linguistic Diversity, September 6-10, Tbilissi, Georgia, Pages 654-660, ISBN 978-9941-13-542-2, Ivane Javakhishvili Tbilisi University Press, Tbilissi, 9/2016.


In this short paper we describe an on-going work, which consists in adding polarity information to entries that are part of the Database of Bavarian Dialects in A ustria (DBÖ). By “polarity information” we mean the positive or negative interpretation a word can carry. The starting point of our study is given by SentiMerge, a lexical resource that encodes polarity i nformation for standard German words on the basis of integration processes performed on four pre-existing polarity lexicons. The lexical information of the entries of Se ntiMerge is encoded using the Ontolex model, which has been developed in the context of the W3C Ontology-Lexica Community Group. The polarity information, on the other hand, is encoded using the MARL ontological model, which has been developed at the Universidad Politécnica de Madrid. The use of such formal representation frameworks is aimed at supporting the publication of the resulting data in the Linked Open Data cloud. In the first phase of our work we focus on headwords describing colour terms, taking also compound words into consideration.


Proceedings_EURALEX2016_Declerck_Dorn_106.pdf (pdf, 416 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence