Visualization of Topic Graph of Weibo Public Opinion Based on Text Mining

Abstract

[Purpose/Significance] Identifying users' concerns from massive comments on Weibo can help the administrative departments to manage the evolution and development trend of public opinion. [Method/Process] Taking Weibo as an example, this paper constructed the topic graph of Weibo public opinion based on the knowledge graph theory and the text mining method. By applying different text similarity, network optimization and text clustering algorithms, the structural characteristics of the graphs were analyzed. [Results/Conclusions] The construction of the topic graph of Weibo public opinion can help managers quickly identify users' concerns. At the same time, it plays an important role in managing users' online texts, predicting the evolution trend, and preventing the diffusion of negative public opinion on social media.

Keywords

• weibo public opinion|text mining|topic graph|visualization