Subjective Relation Identification in Chinese Opinion Mining Based on Sentential Features and Ensemble Classifier
Identifying the subjective relationship is important for opinion mining on product reviews in Chinese. The commonly used identification methods adopt classifiers as the identifier. However, it is difficult to maintain high accuracy and high recall rate simultaneously for the instability of exciting classification method. Motivated by this, we present a method based on sentential features and ensemble classifier to identify the subjective relation from product reviews in Chinese. This method derives sentential futures of candidate feature-opinion pairs from text besides traditional features such as lexical, part of speech, semantic and positional features, and builds sub-classifiers using these features. Then it builds an ensemble classifier with weighted voting mechanism to identify the subjective relationship between feature-opinion pairs. Extensive studies on corpus of book and phone reviews in Chinese demonstrate that the introduction of sentential feature could improve the recall rate of classifier, and a weighted ensemble classifier also could achieve the higher value of F-measure with the trade-off between the accuracy and recall rate of sub-classifiers.
opinion mining subjective relation identification sentential features ensemble classifier
Lijun Shi Jing Zhang Xuegang Hu
School of computer and information Hefei University of Technology Hefei China
国际会议
成都
英文
450-455
2010-07-07（万方平台首次上网日期，不代表论文的发表时间）