Constructing Method of Public Opinion Knowledge Graph with Online News Comments

Abstract

Recently, online news comments research have become a hot issue in Public Emergency Management. This paper proposed a method to construct a knowledge graph of public opinion with online news comments, which is use for the decision-maker master the online public opinion quickly and directly. Firstly, established the object recognition rules of the comments by analyzing the random network structure of the high frequency words in the comments. Secondly, introduced the opinion information structure and the method that extract the structure from comment, which based on CRF models. According to the generated opinion information structure, the comments opinions were extracted and fused. Thereby, the knowledge graph constructed automatically. It had a significance value for information organization and management in public emergency management.