One ofthe richest verb resources currently available forEnglish is VerbNet (Kipper et al., 2000 Kipper, 2005). They dis-play a wide range of syntactic-semantic behaviour,expressing the semantics of an event as well as rela-tional information among its participants (Jackend-off, 1972 Gruber, 1976 Levin, 1993, inter alia).Lexical resources which capture the variabil-ity of verbs are instrumental for many NaturalLanguage Processing (NLP) applications. Playing a key role in conveying the meaning of asentence, verbs are famously complex. Our results show that the pro-posed cross-lingual transfer approach setsnew state-of-the-art verb classification per-formance across all six target languagesexplored in this work. A standard cluster-ing algorithm is then run on top of theVerbNet-specialised representations, usingvector dimensions as features for learningverb classes. ![]() Our method uses cross-lingual trans-lation pairs to tie each of the six target lan-guages into a bilingual vector space withEnglish, jointly specialising the representa-tions to encode the relational informationfrom English VerbNet. To thebest of our knowledge, this is the first studywhich demonstrates how the architecturesfor learning word embeddings can be ap-plied to this challenging syntactic-semantictask. In this work, we propose a novelcross-lingual transfer method for inducingVerbNets for multiple languages. Ivan Vuli´c, Nikola Mrkši´c and Anna Korhonen Language Technology Lab, University of Cambridge, UK Dialogue Systems Group, University of Cambridge, UK approaches to automatic VerbNet-style verb classification are heavily de-pendent on feature engineering and there-fore limited to languages with mature NLPpipelines. Our results show that the proposed cross-lingual transfer approach sets new state-of-the-art verb classification performance across all six target languages explored in this work.ĬCross-Lingual Induction and Transfer of Verb ClassesBased on Word Vector Space Specialisation A standard clustering algorithm is then run on top of the VerbNet-specialised representations, using vector dimensions as features for learning verb classes. Our method uses cross-lingual translation pairs to tie each of the six target languages into a bilingual vector space with English, jointly specialising the representations to encode the relational information from English VerbNet. ![]() To the best of our knowledge, this is the first study which demonstrates how the architectures for learning word embeddings can be applied to this challenging syntactic-semantic task. In this work, we propose a novel cross-lingual transfer method for inducing VerbNets for multiple languages. Existing approaches to automatic VerbNet-style verb classification are heavily dependent on feature engineering and therefore limited to languages with mature NLP pipelines.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |