Incorporating Language Patterns and Domain Knowledge into Feature-Opinion Extraction

摘要: We present a hybrid method for aspect-based sentiment analysis of Chinese restaurant reviews. Two main components are employed so as to extract feature-opinion pairs in the proposed method: domain independent language patterns found and lexical base built The focus on general knowledge which is implicit contained Chinese, thus can be used directly by other domains without any modification. base, hand, targets particular characteristics given acts plug-in part our prototype system, does not affect portability when applying approach practice. Empirical evaluation shows that performs well it gain progressive result each component takes into effective.