Measuring the effect of discourse structure on sentiment analysis

Authors:
Baptiste Chardon;Farah Benamara;Yannick Mathieu;Vladimir Popescu;Nicholas Asher
Affiliations:
IRIT-CNRS, Toulouse University, France,Synapse Développement, Toulouse, France;IRIT-CNRS, Toulouse University, France;LLF-CNRS, Paris 7 University, France;IRIT-CNRS, Toulouse University, France;IRIT-CNRS, Toulouse University, France
Venue:
CICLing'13 Proceedings of the 14th international conference on Computational Linguistics and Intelligent Text Processing - Volume 2
Year:
2013

Citing 7
Cited 0

Building a discourse-tagged corpus in the framework of Rhetorical Structure Theory

SIGDIAL '01 Proceedings of the Second SIGdial Workshop on Discourse and Dialogue - Volume 16
Sentence and expression level annotation of opinions in user-generated discourse

ACL '10 Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics
A probabilistic rating inference framework for mining user preferences from reviews

World Wide Web
Discourse-level relations for opinion analysis

Discourse-level relations for opinion analysis
Polarity analysis of texts using discourse structure

Proceedings of the 20th ACM international conference on Information and knowledge management
Discourse Structure and Sentiment

ICDMW '11 Proceedings of the 2011 IEEE 11th International Conference on Data Mining Workshops
Unsupervised discovery of discourse relations for eliminating intra-sentence polarity ambiguities

EMNLP '11 Proceedings of the Conference on Empirical Methods in Natural Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The aim of this paper is twofold: measuring the effect of discourse structure when assessing the overall opinion of a document and analyzing to what extent these effects depend on the corpus genre. Using Segmented Discourse Representation Theory as our formal framework, we propose several strategies to compute the overall rating. Our results show that discourse-based strategies lead to better scores in terms of accuracy and Pearson's correlation than state-of-the-art approaches.