Sentiment analysis of Chinese microblogging based on sentiment ontology: a case study of ‘7.23 Wenzhou Train Collision’

Sentiment analysis of microblogging texts can facilitate both organisations’ public opinion monitoring and governments’ response strategies development. Nevertheless, most of the existing analysis methods are conducted on Twitter, lacking of sentiment analysis of Chinese microblogging Weibo, and they generally rely on a large number of manually annotated training or machine learning to perform sentiment classification, yielding with difficulties in application. This paper addresses these problems and employs a sentiment ontology model to examine sentiment analysis of Chinese microblogging. We conduct a sentiment analysis of all public microblogging posts about ‘7.23 Wenzhou Train Collision’ broadcasted by Sina microblogging users between 23 July and 1 August 2011. For every day in this time period, we first extract eight dimensions of sentiment expect, joy, love, surprise, anxiety, sorrow, angry, and hate, and then build fuzzy sentiment ontology based on HowNet and semantic similarity for sentiment analysis; we also establish computing methods of influence and sentiment of microblogging texts; and we finally explore the change of public sentiment after ‘7.23 Wenzhou Train Collision’. The results show that the established sentiment analysis method has excellent application, and the change of different emotional values can reflect the success or failure of guiding the public opinion by the government.