An experimental research on sentiment classification of Chinese reviews by semantic orientation method

In recent years, the advent of the Internet, and the widespread use of advanced information technologies in general, have resulted in a surge of information that is freely available online, for example the customerspsila reviews about products. Sentiment classification is just a new technology concerned whether the opinion is positive or negative that reviews online express. There are two types of method in sentiment classification: semantic orientation and machine learning. In this study, we design an experimental research on horror movie reviews. The target is to test whether further classification of the products could improve the accuracy of sentiment classification by semantic orientation method. At last, experimental results of online movie reviews will be discussed which will give some new ideas about sentiment classification in Chinese.