Projecting Embeddings for Domain Adaption : Joint Modeling of Sentiment Analysis in Diverse Domains
Barnes, Jeremy; Klinger, Roman; Schulte im Walde, Sabine (2018): Projecting Embeddings for Domain Adaption : Joint Modeling of Sentiment Analysis in Diverse Domains, in: Emily M. Bender, Leon Derczynski, Pierre Isabelle, u. a. (Hrsg.), Proceedings of the 27th International Conference on Computational Linguistics, Association for Computational Linguistics, S. 818–830.
Details
- Fakultät/Lehrstuhl
- Autorin(nen)/Autor(en)
- Barnes, JeremySchulte im Walde, Sabine
- Titel des Sammelwerks
- Proceedings of the 27th International Conference on Computational Linguistics
- Herausgeberin(nen)/Herausgeber des Sammelwerks
- Bender, Emily M.Derczynski, LeonIsabelle, Pierre
- Konferenz
- 27th International Conference on Computational Linguistics, Santa Fe, New Mexico, USA
- Verlagsangabe
- Association for Computational Linguistics
- Erscheinungsjahr
- 2018
- Sprache(n)
- Englisch
- URL
- Abstract
Domain adaptation for sentiment analysis is challenging due to the fact that supervised classifiers are very sensitive to changes in domain. The two most prominent approaches to this problem are structural correspondence learning and autoencoders. However, they either require long training times or suffer greatly on highly divergent domains. Inspired by recent advances in cross-lingual sentiment analysis, we provide a novel perspective and cast the domain adaptation problem as an embedding projection task. Our model takes as input two mono-domain embedding spaces and learns to project them to a bi-domain space, which is jointly optimized to (1) project across domains and to (2) predict sentiment. We perform domain adaptation experiments on 20 source-target domain pairs for sentiment classification and report novel state-of-the-art results on 11 domain pairs, including the Amazon domain adaptation datasets and SemEval 2013 and 2016 datasets. Our analysis shows that our model performs comparably to state-of-the-art approaches on domains that are similar, while performing significantly better on highly divergent domains. Our code is available at https://github.com/jbarnesspain/domain_blse
- GND-Schlagwort(e)
- Maschinelles LernenSprachverarbeitung
- Freie Schlagwort(e)
- Projecting Embeddings
- DDC-Klassifikation
- 004 Computer science
- RVK-Klassifikation
- ST 306
- Peer Reviewed
- ja
- Internationale Verbreitung
- ja
- Open-Access-Zeitschrift
- ja
- Dokumenttyp
- Konferenzveröffentlichung
- URI
- Datum der Freischaltung
- 14.03.2024