Annotation of regular polysemy and underspecification

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

We present the result of an annotation task on regular polysemy for a series of seman- tic classes or dot types in English, Dan- ish and Spanish. This article describes the annotation process, the results in terms of inter-encoder agreement, and the sense distributions obtained with two methods: majority voting with a theory-compliant backoff strategy, and MACE, an unsuper- vised system to choose the most likely sense from all the annotations.
Original languageEnglish
Title of host publicationProceedings from the 51st annual meeting in Association for Computational Linguistics
Place of PublicationSofia (Bulgaria)
PublisherAssociation for Computational Linguistics
Publication date2013
Pages725-730
ISBN (Electronic)978 1627489775
Publication statusPublished - 2013

ID: 113412048