A Bayesian Network Model for predicting data breaches caused by insiders of a health care organization
暂无分享,去创建一个
In the Netherlands organizations are required by law to protect personal data. Since January 2016 they are also required to report breaches of security leading to (a considerable likelihood of) serious adverse effects on the protection of personal data to the Dutch data protection authority. In the health care sector medical data, which is extra sensitive, is processed and therefore security is even more important. Data breaches are, in this sector, mostly caused by insiders who have malicious intentions or make mistakes. For organizations it is hard to protect themselves effectively against insider threats and make sure that data breaches do not occur. To help organizations determine whether a data breach is likely to occur Bayesian networks can be used. During this research a model has been developed that combines prior indicators of a data breach and measures taken by an organization to predict the probability of a data breach in a health care. The model combines malicious and accidental insider threats posed by a group of insiders. When changing the observations the best combination of measures to minimize the probability of a data breach given certain prior indicators can be identified.