Unsupervised subjectivity-lexicon generation based on vector space model for multi-dimensional opinion analysis in blogosphere

Authors:
Hsieh-Wei Chen;Kuan-Rong Lee;Hsun-Hui Huang;Yaw-Huang Kuo
Affiliations:
Lab, Dept. of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, ROC;Dept. of Information Engineering, Kun Shan University, Yung-Kang, Tainan, Taiwan, ROC;Lab, Dept. of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, ROC;Lab, Dept. of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, ROC
Venue:
ICIC'10 Proceedings of the 6th international conference on Advanced intelligent computing theories and applications: intelligent computing
Year:
2010

Citing 16
Cited 0

An algorithm for pronominal anaphora resolution

Computational Linguistics
Unsupervised Feature Selection Using Feature Similarity

IEEE Transactions on Pattern Analysis and Machine Intelligence
A centering approach to pronouns

ACL '87 Proceedings of the 25th annual meeting on Association for Computational Linguistics
Thumbs up or thumbs down?: semantic orientation applied to unsupervised classification of reviews

ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
Thumbs up?: sentiment classification using machine learning techniques

EMNLP '02 Proceedings of the ACL-02 conference on Empirical methods in natural language processing - Volume 10
Learning subjective nouns using extraction pattern bootstrapping

CONLL '03 Proceedings of the seventh conference on Natural language learning at HLT-NAACL 2003 - Volume 4
Learning extraction patterns for subjective expressions

EMNLP '03 Proceedings of the 2003 conference on Empirical methods in natural language processing
Topic sentiment mixture: modeling facets and opinions in weblogs

Proceedings of the 16th international conference on World Wide Web
Opinion retrieval from blogs

Proceedings of the sixteenth ACM conference on Conference on information and knowledge management
Aspect Summarization from Blogsphere for Social Study

ICDMW '07 Proceedings of the Seventh IEEE International Conference on Data Mining Workshops
Opinion Mining and Sentiment Analysis

Foundations and Trends in Information Retrieval
An effective statistical approach to blog post opinion retrieval

Proceedings of the 17th ACM conference on Information and knowledge management
Rated aspect summarization of short comments

Proceedings of the 18th international conference on World wide web
A survey on sentiment detection of reviews

Expert Systems with Applications: An International Journal
The Stanford typed dependencies representation

CrossParser '08 Coling 2008: Proceedings of the workshop on Cross-Framework and Cross-Domain Parser Evaluation
Feature subsumption for opinion analysis

EMNLP '06 Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an unsupervised framework to generate a vectorspace-modeled subjectivity-lexicon for multi-dimensional opinion mining and sentiment analysis, such as criticism analysis, for which the traditional polarity analysis alone is not adequate. The framework consists of four major steps: first, creating a dataset by crawling blog posts of fiction reviews; secondly, creating a "subjectivity-term to object" matrix, with each subjectivity-term being modeled as a dimension of a vector space; thirdly, feature-transforming each subjectivity-term into the new feature-space to create the final multidimensional subjectivity-lexicon (MDSL); and fourthly, using the generated MDSL for opinion analysis. In the experiments, it shows that the improvement by the feature transform can be up to 31% in terms of the entropy of features. In addition, the subjectivity-terms and objects are also successfully and reasonably clustered in the demonstration of fiction review (literary criticism) analysis.