Cross-lingual Model Transfer Using Feature Representation Projection

Autor:	Mikhail Kozhevnikov, Ivan Titov
Rok vydání:	2014
Předmět:	Annotation Transfer (group theory) Cross lingual Semantic role labeling Computer science Feature (computer vision) business.industry Representation (systemics) Pattern recognition Artificial intelligence Projection (set theory) business Task (project management)
Zdroj:	ACL (2)
DOI:	10.3115/v1/p14-2095
Popis:	We propose a novel approach to crosslingual model transfer based on feature representation projection. First, a compact feature representation relevant for the task in question is constructed for either language independently and then the mapping between the two representations is determined using parallel data. The target instance can then be mapped into the source-side feature representation using the derived mapping and handled directly by the source-side model. This approach displays competitive performance on model transfer for semantic role labeling when compared to direct model transfer and annotation projection and suggests interesting directions for further research.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::d5168dd8edb6f10516296c29efff1241 https://doi.org/10.3115/v1/p14-2095 Zobrazit plný text záznamu