Abstract: Internet news integrates advantages of fast, real and massive information, and has become an indispensable supervisory force in the security market. Internet news can dig deep into the hidden inside information of listed companies and avoid the collapse of stock prices caused by the continuous hoarding of bad news. Therefore, the unstructured data associated with news information of the securities are chooses. The data from UQER quantitative trading platform are obtained. The sample period is from January 1, 2014 to December 31, 2017. The Heatindex and the Daynumber variables are used to measure the media buzz degree of a stock. The risk of stock price collapse is measured by CRASH variable, NCSKEW variable and DUVOL variable. We aim to explore the influence of media buzz degree on the risk of future stock price collapse and the role of corporate transparency's moderating. The empirical analysis shows that the higher media buzz degree represents the lower the risk of stock price collapse. Media buzz degree can suppress significantly the corporate stock price crash risk, specifically in the low transparency firms such as state-owned enterprises, low institutional ownership and low transparency of financial reporting. The findings indicate that we should correctly guide all kinds of financial news reports in the process of securities market supervision system is increasingly perfect and strengthen the supervisory role of news media so that reduce the stock price crash risk and protect the interests of the investors. This research supports the demonstration of effective media governance and makes up for the deficiency of single source on media attention.

Key words: internet finance, inclusive finance, media buzz degree, stock price crash, corporate transparency