A Template-Based Approach to Eztract Product Features and Sentiment Words
This paper proposed a new algorithm to extract product features and the corresponding sentiment words from Chinese product reviews. The algorithm is a departure from previous work in that: 1) it utilizes the relationship between product features and the corresponding sentiment words to extract the two kinds of words mutually and iteratively; 2) it is domain-independent. Without the use of any domain related training corpus and given several domain related seed words, the algorithm can be applied to many different domains. Our experiment results show that the algorithm gained good and stable performance in different domains.
product reviews analysis templates of POS tags product features sentiment words
Wenjing Zhao Yanquan Zhou
Beijing University of Posts and Telecommunications
国际会议
大连
英文
1-5
2009-09-24（万方平台首次上网日期，不代表论文的发表时间）