摘要
Many organizations have datasets which contain a high volume of personal data on individuals,e.g.,health data.Even without a name or address,persons can be identified based on the details(variables)on the dataset.This is an important issue for big data holders such as public sector organizations(e.g.,Public Health Organizations)and social media companies.This paper looks at how individuals can be identified from big data using a mathematical approach and how to apply this mathematical solution to prevent accidental disclosure of a person’s details.The mathematical concept is known as the“Identity Correlation Approach”(ICA)and demonstrates how an individual can be identified without a name or address using a unique set of characteristics(variables).Secondly,having identified the individual person,it shows how a solution can be put in place to prevent accidental disclosure of the personal details.Thirdly,how to store data such that accidental leaks of the datasets do not lead to the disclosure of the personal details to unauthorized users.
