A study on Web security incidents in China by analyzing vulnerability disclosure platforms

Understanding the nature of a country's World Wide Web security can allow analysts to evaluate the security awareness of local organizations, the technology employed by researchers, and the defense capabilities of the whole country. In this paper, we put forward a new framework to evaluate the security situation in China with real vulnerability disclosure platforms. The focus of this research is to analyze the current situation of Chinese websites using 57,112 Web vulnerability incidents submitted by 5371 researchers from 2012 to 2015. The dataset is distributed into four types of organizations, including listed companies, government institutions, educational institutions, and startups. We present an approach, based on machine learning and natural language processing technologies, to classify the vulnerability type for each incident. Furthermore, our experimental results show that the vulnerability distribution and response speed toward important issues are so different among the four types of organizations that researchers at various levels of experience begin to take part in submitting vulnerabilities to public disclosure platforms. Based on the results, we propose security some best-practices for organizations and show that the security situation of Chinese websites has changed quickly in the last three years but is still facing several big problems.