Abstract: Using web crawler, we crawled users’ comment data on two hot events, M_S5.2 earthquake in Beiliu and M_S5.2 earthquake in Jingxi released by CCTV News microblog, the visualized word cloud picture of each individual earthquake event and the statistical table of the top ten hot frequency words were produced using data processing techniques such as cleaning, word segmentation, removing stop-words and word cloud statistics. Through mining and analyzing the public opinion information of the word cloud chart and the hot frequency word statistical table, it is found that in two hot issues about earthquake, the main content of microblog users' comments was to express their wishes for peace; participation of users in the former are more engaged; compared with the latter, the former has a wider range of users.