Abstract
Our aim is to develop principled methods for sense clustering which can make existing lexi-cal resources practically useful in NLP - not too fine-grained to be operational and yet fine-grained enough to be worth the trouble. Where traditional dictionaries have a highly structured sense inventory typically describing the vocabulary by means of main- and subsenses, wordnets are generally fine-grained and un-structured. We present a series of clustering and annotation experiments with 10 of the most polysemous nouns in Danish. We com-bine the structured information of a traditional Danish dictionary with the ontological types found in the Danish wordnet, DanNet. This constellation enables us to automatically clus-ter senses in a principled way and improve in-ter-annotator agreement and wsd performance.
Original language | English |
---|---|
Title of host publication | Proceedings of Global WordNet Conference 2018 |
Number of pages | 6 |
Place of Publication | Singapore |
Publisher | Global WordNet Association |
Publication date | 2018 |
ISBN (Electronic) | 978-981-11-7087-4 |
Publication status | Published - 2018 |
Event | Global WordNet Conference - Singapore, Singapore Duration: 8 Jan 2018 → 12 Jan 2018 |
Conference
Conference | Global WordNet Conference |
---|---|
Country/Territory | Singapore |
City | Singapore |
Period | 08/01/2018 → 12/01/2018 |