Publication Type

Conference Proceeding Article

Publication Date

11-2016

Abstract

In this paper, we study cross-domain sentimentclassification with neural network architectures.We borrow the idea from StructuralCorrespondence Learning and use two auxiliarytasks to help induce a sentence embeddingthat supposedly works well across domains forsentiment classification. We also propose tojointly learn this sentence embedding togetherwith the sentiment classifier itself. Experimentresults demonstrate that our proposedjoint model outperforms several state-of-the artmethods on five benchmark datasets.

Discipline

Databases and Information Systems | Software Engineering

Research Areas

Data Management and Analytics

Publication

EMNLP 2016: Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing: Austin, Texas, 2016 November 1-25

Publisher

Association for Computational Linguistics

City or Country

Stroudsburg, USA

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS