%A Zhang Zhiwu %T Sentiment Analysis of Product Reviews by means of Cross-domain Transfer Learning %0 Journal Article %D 2013 %J Data Analysis and Knowledge Discovery %R 10.11925/infotech.1003-3513.2013.06.08 %P 49-54 %V %N 6 %U {https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/abstract/article_3749.shtml} %8 2013-06-25 %X Aiming at the problem of sentiment analysis of incomplete product reviews data, this paper proposes a cross-domain sentiment analysis method based on spectral clustering and transfer learning. With the help of domain-independent words as a bridge, using spectral clustering algorithm to align domain-specific words from different domains into unified clusters, it can reduce the gap between domain-specific words of the two domains, and can improve the accuracy of sentiment classifiers in the target domain. Experiments studies are carried out to show the efficiency and superiority of the proposed approach in solving the problem of cross-domain sentiment analysis of product reviews.