SODA

Discovering fine-grained sentiment with latent variable structured prediction models

Täckström, Oscar and McDonald, Ryan (2011) Discovering fine-grained sentiment with latent variable structured prediction models. In: The 33rd European Conference on Information Retrieval, 18-21 Apr 2011, Dublin, Ireland. (In Press)

[img]
Preview
PDF - Accepted Version
416Kb

Abstract

In this paper we investigate the use of latent variable structured prediction models for fine-grained sentiment analysis in the common situation where only coarse-grained supervision is available. Specifically, we show how sentence-level sentiment labels can be effectively learned from document-level supervision using hidden conditional random fields (HCRFs). Experiments show that this technique reduces sentence classification errors by 22% relative to using a lexicon and 13% relative to machine-learning baselines.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:Sentiment analysis, Latent variables, Structured conditional models
ID Code:4061
Deposited By:Oscar Tackström
Deposited On:20 Jan 2011 10:01
Last Modified:20 Jan 2011 10:01

Repository Staff Only: item control page