Papers by Thomas Eckart
Typical Sentences as a Resource for Valence (2020.lrec-1)
Copied to clipboard
| Challenge: | Verb valence information can be derived from corpora by using subcorpora of typical sentences that are constructed in a language independent manner based on frequent POS structures. |
| Approach: | They propose a method which allows one to 'strip' the POS-pattern of a sentence with various types of adjuncts down to those items representing the valence frame instantiated by the verb in question. |
| Outcome: | The proposed approach can be used to construct valence pairs for a bilingual polyval dictionary using the Norwegian valance dictionary NorVal. |
Preparation and Usage of Xhosa Lexicographical Data for a Multilingual, Federated Environment (L18-1)
Copied to clipboard
| Challenge: | lexicographical data is often difficult to find for less resourced languages . Xhosa is a popular language in south africa, but it is often suboptimal for many languages despite its multilingual nature . |
| Approach: | They propose a new source of lexicographical data for Xhosa, a language spoken by 8 million speakers. |
| Outcome: | The proposed model can be used in multilingual and federated environments and is extensible to other languages. |