Building an Anonymization Pipeline

Index

A

• accidental (inadvertent) attack, Safe Data, Five Safes in Practice
• accountability, as tenet of governance, A Practical Risk-Management Framework, Safe Settings, Use Cases, Safe Pooling
• acquaintances
  • accidental re-identification by, Five Safes in Practice
  • among adversaries, Population to sample (acquaintance)
  • among data recipients, Safe People, Acquaintances, Five Safes in Practice
  • indirect identifiers known by, Categories of information, A First Step to Anonymization, Identifiability Spectrum Revisited
• adaptive models, Algorithms Failing on Trust
• addressability, Dimensions of Data Privacy, Addressability, Identifiability
• adversaries
  • ability to verify re-identifications, Overall Identifiability
  • acquaintances as, Population to sample (acquaintance)
  • categories of, Safe People
  • data recipients considered as, Anonymization as Data Protection-Anonymization as Data Protection, Safe People, Recipient trust
  • defined population and, Defined Population, Defined Population
  • differential privacy and, Differential privacy
  • direction of matching used by, Direction of Matching-Population to sample (acquaintance)
  • “learning something new” criteria for, Learning something new
  • power of, Cross-sectional data, Longitudinal or panel data
  • risk matrix for, Risk matrix-Risk matrix
• aggregator portals, Additional Sources of Data
• AIML (artificial intelligence and machine learning), Techniques of AIML-Technical Challenges
  • algorithm development for, Synthetic Data
  • data monitoring of, Data Monitoring
  • responsible use of, Principles ...