论文部分内容阅读
Sentiment analysis of online reviews and other user generated content is an important research problem for its wide range of applications.In this paper,we propose a feature-based vector model and a novel weighting algorithm for sentiment analysis of Chinese product reviews.Specifically,an opinionated document is modeled by a set of feature-based vectors and corresponding weights.Different from previous work,our model considers modifying relationships between words and contains rich sentiment strength descriptions which are represented by adverbs of degree and punctuations.Dependency parsing is applied to construct the feature vectors.A novel feature weighting algorithm is proposed for supervised sentiment classification based on rich sentiment strength related information.The experimental results demonstrate the effectiveness of the proposed method compared with a state of the art method using term level weighting algorithms.
Sentiment analysis of online reviews and other user generated content is an important research problem for its wide range of applications. In this paper, we propose a feature-based vector model and a novel weighting algorithm for sentiment analysis of Chinese product reviews. Specifically, an opinionated document is modeled by a set of feature-based vectors and corresponding weights. different from previous work, our model considers modifying relationships between words and contains rich sentiment strength descriptions which are represented by adverbs of degree and punctuations. Dependency parsing is applied to construct the feature vectors. A novel feature weighting algorithm is proposed for supervised sentiment classification based on rich sentiment strength related information information. The experimental results demonstrate the effectiveness of the proposed method compared with a state of the art method using term level weighting algorithms.