A template-based approach to extract 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.