Potential and Limitations of Cross-Domain Sentiment Classification

Jan Milan Deriu, Martin Weilenmann, Dirk Von Gruenigen, Mark Cieliebak


Abstract
In this paper we investigate the cross-domain performance of a current state-of-the-art sentiment analysis systems. For this purpose we train a convolutional neural network (CNN) on data from different domains and evaluate its performance on other domains. Furthermore, we evaluate the usefulness of combining a large amount of different smaller annotated corpora to a large corpus. Our results show that more sophisticated approaches are required to train a system that works equally well on various domains.
Anthology ID:
W17-1103
Volume:
Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media
Month:
April
Year:
2017
Address:
Valencia, Spain
Editors:
Lun-Wei Ku, Cheng-Te Li
Venue:
SocialNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
17–24
Language:
URL:
https://aclanthology.org/W17-1103
DOI:
10.18653/v1/W17-1103
Bibkey:
Cite (ACL):
Jan Milan Deriu, Martin Weilenmann, Dirk Von Gruenigen, and Mark Cieliebak. 2017. Potential and Limitations of Cross-Domain Sentiment Classification. In Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media, pages 17–24, Valencia, Spain. Association for Computational Linguistics.
Cite (Informal):
Potential and Limitations of Cross-Domain Sentiment Classification (Deriu et al., SocialNLP 2017)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/W17-1103.pdf