TY - GEN
T1 - Instance weighting for domain adaptation in NLP
AU - Jiang, Jing
AU - Zhai, Cheng Xiang
N1 - Funding Information:
Nadie mejor situado que Quevedo —que, como es sabido, desde poco después del acceso al poder de Felipe IV y Olivares residió de forma bastante continuada en Madrid— para observar la proliferación de cortesanos descreídos, para oír el insistente remoquete «Los hombres mueren como las bestias», y hasta para leer los escritos clandestinos a los que se refería la denuncia de Poza. Es cierto que no tenemos tes-timonios personales del propio Quevedo al respecto, pero el caso de López de Vega, tan semejante, nos invita a pensar que en el origen del retrato del cortesano ateísta que Quevedo ofrece en Providencia de Dios i, y del diálogo que mantiene con él —o más bien la reconvención que le dirige—, estuvo una observación directa de lo que estaba ocurriendo en la sociedad madrileña en la que se movía.
PY - 2007
Y1 - 2007
N2 - Domain adaptation is an important problem in natural language processing (NLP) due to the lack of labeled data in novel domains. In this paper, we study the domain adaptation problem from the instance weighting perspective. We formally analyze and characterize the domain adaptation problem from a distributional view, and show that there are two distinct needs for adaptation, corresponding to the different distributions of instances and classification functions in the source and the target domains. We then propose a general instance weighting framework for domain adaptation. Our empirical results on three NLP tasks show that incorporating and exploiting more information from the target domain through instance weighting is effective.
AB - Domain adaptation is an important problem in natural language processing (NLP) due to the lack of labeled data in novel domains. In this paper, we study the domain adaptation problem from the instance weighting perspective. We formally analyze and characterize the domain adaptation problem from a distributional view, and show that there are two distinct needs for adaptation, corresponding to the different distributions of instances and classification functions in the source and the target domains. We then propose a general instance weighting framework for domain adaptation. Our empirical results on three NLP tasks show that incorporating and exploiting more information from the target domain through instance weighting is effective.
UR - http://www.scopus.com/inward/record.url?scp=84860538689&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84860538689&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84860538689
SN - 9781932432862
T3 - ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics
SP - 264
EP - 271
BT - ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics
T2 - 45th Annual Meeting of the Association for Computational Linguistics, ACL 2007
Y2 - 23 June 2007 through 30 June 2007
ER -