Text sentiment classification aims to automatically classify subjective documents into different sentiment-oriented categories (e.g. positive/negative). Given the high dimensionality of features describing documents, how to effectively select the most useful ones, referred to as sentiment-bearing features, with a lack of sentiment class labels is crucial for improving the classification performance. This paper proposes an unsupervised sentiment-bearing feature selection method (USFS), which incorporates sentiment discriminant analysis (SDA) into sentiment strength calculation (SSC). SDA applies traditional linear discriminant analysis (LDA) in an unsupervised manner without losing local sentiment information between documents. We use SSC to calculate the overall sentiment strength for each single feature based on its affinities with some sentiment priors. Experiments, performed using benchmark movie reviews, demonstrated the superior performance of USFS.
Yan LI
Beijing University of Posts and Telecommunications
Zhen QIN
Beijing University of Posts and Telecommunications
Weiran XU
Beijing University of Posts and Telecommunications
Heng JI
Troy
Jun GUO
Beijing University of Posts and Telecommunications
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yan LI, Zhen QIN, Weiran XU, Heng JI, Jun GUO, "Unsupervised Sentiment-Bearing Feature Selection for Document-Level Sentiment Classification" in IEICE TRANSACTIONS on Information,
vol. E96-D, no. 12, pp. 2805-2813, December 2013, doi: 10.1587/transinf.E96.D.2805.
Abstract: Text sentiment classification aims to automatically classify subjective documents into different sentiment-oriented categories (e.g. positive/negative). Given the high dimensionality of features describing documents, how to effectively select the most useful ones, referred to as sentiment-bearing features, with a lack of sentiment class labels is crucial for improving the classification performance. This paper proposes an unsupervised sentiment-bearing feature selection method (USFS), which incorporates sentiment discriminant analysis (SDA) into sentiment strength calculation (SSC). SDA applies traditional linear discriminant analysis (LDA) in an unsupervised manner without losing local sentiment information between documents. We use SSC to calculate the overall sentiment strength for each single feature based on its affinities with some sentiment priors. Experiments, performed using benchmark movie reviews, demonstrated the superior performance of USFS.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E96.D.2805/_p
Copy
@ARTICLE{e96-d_12_2805,
author={Yan LI, Zhen QIN, Weiran XU, Heng JI, Jun GUO, },
journal={IEICE TRANSACTIONS on Information},
title={Unsupervised Sentiment-Bearing Feature Selection for Document-Level Sentiment Classification},
year={2013},
volume={E96-D},
number={12},
pages={2805-2813},
abstract={Text sentiment classification aims to automatically classify subjective documents into different sentiment-oriented categories (e.g. positive/negative). Given the high dimensionality of features describing documents, how to effectively select the most useful ones, referred to as sentiment-bearing features, with a lack of sentiment class labels is crucial for improving the classification performance. This paper proposes an unsupervised sentiment-bearing feature selection method (USFS), which incorporates sentiment discriminant analysis (SDA) into sentiment strength calculation (SSC). SDA applies traditional linear discriminant analysis (LDA) in an unsupervised manner without losing local sentiment information between documents. We use SSC to calculate the overall sentiment strength for each single feature based on its affinities with some sentiment priors. Experiments, performed using benchmark movie reviews, demonstrated the superior performance of USFS.},
keywords={},
doi={10.1587/transinf.E96.D.2805},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - Unsupervised Sentiment-Bearing Feature Selection for Document-Level Sentiment Classification
T2 - IEICE TRANSACTIONS on Information
SP - 2805
EP - 2813
AU - Yan LI
AU - Zhen QIN
AU - Weiran XU
AU - Heng JI
AU - Jun GUO
PY - 2013
DO - 10.1587/transinf.E96.D.2805
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2013
AB - Text sentiment classification aims to automatically classify subjective documents into different sentiment-oriented categories (e.g. positive/negative). Given the high dimensionality of features describing documents, how to effectively select the most useful ones, referred to as sentiment-bearing features, with a lack of sentiment class labels is crucial for improving the classification performance. This paper proposes an unsupervised sentiment-bearing feature selection method (USFS), which incorporates sentiment discriminant analysis (SDA) into sentiment strength calculation (SSC). SDA applies traditional linear discriminant analysis (LDA) in an unsupervised manner without losing local sentiment information between documents. We use SSC to calculate the overall sentiment strength for each single feature based on its affinities with some sentiment priors. Experiments, performed using benchmark movie reviews, demonstrated the superior performance of USFS.
ER -