HITSZ_CITYU: Combine collocation, context words and neighboring sentence sentiment in sentiment adjectives disambiguation

Authors:
Ruifeng Xu;Jun Xu;Chunyu Kit
Affiliations:
Harbin Institute of Technology, Shenzhen Campus, China and City University of Hong Kong, Hong Kong;Harbin Institute of Technology, Shenzhen Campus, China;City University of Hong Kong, Hong Kong
Venue:
SemEval '10 Proceedings of the 5th International Workshop on Semantic Evaluation
Year:
2010

Citing 4
Cited 1

Measuring praise and criticism: Inference of semantic orientation from association

ACM Transactions on Information Systems (TOIS)
Predicting the semantic orientation of adjectives

ACL '98 Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics and Eighth Conference of the European Chapter of the Association for Computational Linguistics
Learning Knowledge from Relevant Webpage for Opinion Analysis

WI-IAT '08 Proceedings of the 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology - Volume 01
Automatic identification of sentiment vocabulary: exploiting low association with known sentiment terms

FeatureEng '05 Proceedings of the ACL Workshop on Feature Engineering for Machine Learning in Natural Language Processing

SemEval-2010 task 18: disambiguating sentiment ambiguous adjectives

Language Resources and Evaluation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents the HIT_CITYU systems in Semeval-2 Task 18, namely, disambiguating sentiment ambiguous adjectives. The baseline system (HITSZ_CITYU_3) incorporates bi-gram and n-gram collocations of sentiment adjectives, and other context words as features in a one-class Support Vector Machine (SVM) classifier. To enhance the baseline system, collocation set expansion and characteristics learning based on word similarity and semisupervised learning are investigated, respectively. The final system (HITSZ_CITYU_1/2) combines collocations, context words and neighboring sentence sentiment in a two-class SVM classifier to determine the polarity of sentiment adjectives. The final systems achieved 0.957 and 0.953 (ranked 1st and 2nd) macro accuracy, and 0.936 and 0.933 (ranked 2nd and 3rd) micro accuracy, respectively.