Joint Topic-Opinion Model for Implicit Feature Extracting

Topic model has been used to extract implicit features yet little concerns have been given to general opinion words, e.g., "Okey" (good). In this paper we present a modified topic model joint topic-opinion model (JTO) for extracting implicit features of opinion words including special and general ones. Our model is based on an extension to standard LDA model by adding an opinion level. This model considers both topics and context of opinion words. Experiments show that JTO provides higher accuracy in implicit features extraction.