Learning representations for sentiment classification using Multi-task framework

Hardik Meisheri, Harshad Khadilkar


Abstract
Most of the existing state of the art sentiment classification techniques involve the use of pre-trained embeddings. This paper postulates a generalized representation that collates training on multiple datasets using a Multi-task learning framework. We incorporate publicly available, pre-trained embeddings with Bidirectional LSTM’s to develop the multi-task model. We validate the representations on an independent test Irony dataset that can contain several sentiments within each sample, with an arbitrary distribution. Our experiments show a significant improvement in results as compared to the available baselines for individual datasets on which independent models are trained. Results also suggest superior performance of the representations generated over Irony dataset.
Anthology ID:
W18-6244
Volume:
Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis
Month:
October
Year:
2018
Address:
Brussels, Belgium
Editors:
Alexandra Balahur, Saif M. Mohammad, Veronique Hoste, Roman Klinger
Venue:
WASSA
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
299–308
Language:
URL:
https://aclanthology.org/W18-6244
DOI:
10.18653/v1/W18-6244
Bibkey:
Cite (ACL):
Hardik Meisheri and Harshad Khadilkar. 2018. Learning representations for sentiment classification using Multi-task framework. In Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, pages 299–308, Brussels, Belgium. Association for Computational Linguistics.
Cite (Informal):
Learning representations for sentiment classification using Multi-task framework (Meisheri & Khadilkar, WASSA 2018)
Copy Citation:
PDF:
https://aclanthology.org/W18-6244.pdf