A system for fine-grained aspect-based sentiment analysis of Chinese

This paper presents a pipeline for aspect-based sentiment analysis of Chinese texts in the automotive domain. The input to the pipeline is a string of Chinese characters; the output is a set of relationships between evaluations and their targets. The main goal is to demonstrate how knowledge about sentence structure can increase the precision, insight value and granularity of the output. We formulate the task of sentiment analysis in two steps, namely unit identification and relation extraction. In unit identification, we identify fairly well-delimited linguistic units which describe features, emotions and evaluations. In relation extraction, we discover the relations between evaluations and their "target" features.