Privacy Protection for Big Data Linking using the Identity Correlation Approach

Privacy protection for big data linking is discussed here in relation to the Central Statistics Office （CSO）, Ireland＇s, big data linking project titled the ＇Structure of Earnings Survey - Administrative Data Project＇ （SESADP）. The result of the project was the creation of datasets and statistical outputs for the years 2011 to 2014 to meet Eurostat＇s annual earnings statistics requirements and the Structure of Earnings Survey （SES） Regulation. Record linking across the Census and various public sector datasets enabled the necessary information to be acquired to meet the Eurostat earnings requirements. However, the risk of statistical disclosure （i.e. identifying an individual on the dataset） is high unless privacy and confidentiality safe-guards are built into the data matching process. This paper looks at the three methods of linking records on big datasets employed on the SESADP, and how to anonymise the data to protect the identity of the individuals, where potentially disclosive variables exist.

A Mathematical Solution to String Matching for Big Data Linking

This paper describes how data records can be matched across large datasets using a technique called the Identity Correlation Approach （ICA）. The ICA technique is then compared with a string matching exercise. Both the string matching exercise and the ICA technique were employed for a big data project carried out by the CSO. The project was called the SESADP （Structure of Earnings Survey Administrative Data Project） and involved linking the Irish Census dataset 2011 to a large Public Sector Dataset. The ICA technique provides a mathematical tool to link the datasets and the matching rate for an exact match can be calculated before the matching process begins. Based on the number of variables and the size of the population, the matching rate is calculated in the ICA approach from the MRUI （Matching Rate for Unique Identifier） formula, and false positives are eliminated. No string matching is used in the ICA, therefore names are not required on the dataset, making the data more secure ＆ ensuring confidentiality. The SESADP Project was highly successful using the ICA technique. A comparison of the results using a string matching exercise for the SESADP and the ICA are discussed here.

Cyber Security-Protecting Personal Data

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.