When collecting data about people, researchers don’t want to accidentally cause harm, or risk sharing too much information.
They try to reduce the amount of personal data in their data sets. By minimising personal data, they minimise the dangers.
But if it's not done properly, can this be undone? Is there a data disaster waiting to happen?
Read and unmask the 'anonymised' job applications
Decode browsing history to find the secretive 'Stardust Streams'
Compare a medical dataset with 'open' data to identify patients
Run Python code to automate the re-identification process