Data Privacy

• Main Author: Torra, Vicenç
• Format: eBook
• Language: English
• Published: Cham Springer International Publishing AG 2017
• Edition: 1
• Subjects: Artificial intelligence
• ISBN: 9783319573564; 331957356X

Table of Contents:

• 5.1.1 A Boolean or Measurable Condition -- 5.2 Attribute Disclosure -- 5.2.1 Attribute Disclosure for a Numerical Variable -- 5.2.2 Attribute Disclosure for a Categorical Variable -- 5.3 Identity Disclosure -- 5.3.1 An Scenario for Identity Disclosure -- 5.3.2 Measures for Identity Disclosure -- 5.3.3 Uniqueness -- 5.3.4 Reidentification -- 5.3.5 The Worst-Case Scenario -- 5.4 Matching and Integration: A Database Based Approach -- 5.4.1 Heterogenous Distributed Databases -- 5.4.2 Data Integration -- 5.4.3 Schema Matching -- 5.4.4 Data Matching -- 5.4.5 Preprocessing -- 5.4.6 Indexing and Blocking -- 5.4.7 Record Pair Comparison: Distances and Similarities -- 5.4.8 Classification of Record Pairs -- 5.5 Probabilistic Record Linkage -- 5.5.1 Alternative Expressions for Decision Rules -- 5.5.2 Computation of Rp(a,b) -- 5.5.3 Estimation of the Probabilities -- 5.5.4 Extensions for Computing Probabilities -- 5.5.5 Final Notes -- 5.6 Distance-Based Record Linkage -- 5.6.1 Weighted Distances -- 5.6.2 Distance and Normalization -- 5.6.3 Parameter Determination for Record Linkage -- 5.7 Record Linkage Without Common Variables -- 5.8 k-Anonymity and Other Boolean Conditions for Identity Disclosure -- 5.8.1 k-Anonymity and Anonymity Sets: k-Confusion -- 5.8.2 k-Anonymity and Attribute Disclosure: Attacks -- 5.8.3 Other Related Approaches to k-Anonymity -- 5.9 Discussion on Record Linkage -- 5.9.1 Formalization of Reidentification Algorithms -- 5.9.2 Comparison of Record Linkage Algorithms -- 5.9.3 Disclosure Risk and Big Data -- 5.9.4 Some Guidelines and Research Issues -- 6 Masking Methods -- 6.1 Perturbative Methods -- 6.1.1 Data and Rank Swapping -- 6.1.2 Microaggregation -- 6.1.3 Additive and Multiplicative Noise -- 6.1.4 PRAM -- 6.1.5 Lossy Compression and Other Transform-Based Method -- 6.2 Non-perturbative Methods -- 6.2.1 Generalization and Recoding
• 6.2.2 Suppression -- 6.3 Synthetic Data Generators -- 6.4 Masking Methods and k-anonymity -- 6.4.1 Mondrian -- 6.4.2 Algorithms for k-anonymity: Variants and Big Data -- 6.5 Data Protection Procedures for Constrained Data -- 6.5.1 Types of Constraints -- 6.6 Masking Methods and Big Data -- 7 Information Loss: Evaluation and Measures -- 7.1 Generic Versus Specific Information Loss -- 7.2 Information Loss Measures -- 7.3 Generic Information Loss Measures -- 7.3.1 Numerical Data -- 7.3.2 Categorical Data -- 7.4 Specific Information Loss -- 7.4.1 Classification-Based Information Loss -- 7.4.2 Regression-Based Information Loss -- 7.4.3 Clustering-Based Information Loss -- 7.5 Information Loss and Big Data -- 8 Selection of Masking Methods -- 8.1 Aggregation: A Score -- 8.2 Visualization: R-U Maps -- 8.3 Optimization and Postmasking -- 9 Conclusions -- Index -- Index
• Intro -- Preface -- Organization -- How to Use This book -- Acknowledgements -- Contents -- 1 Introduction -- 1.1 Motivations for Data Privacy -- 1.2 Privacy and Society -- 1.3 Terminology -- 1.3.1 The Framework -- 1.3.2 Anonymity and Unlinkability -- 1.3.3 Disclosure -- 1.3.4 Undetectability and Unobservability -- 1.3.5 Pseudonyms and Identity -- 1.3.6 Transparency -- 1.4 Privacy and Disclosure -- 1.5 Privacy by Design -- 2 Machine and Statistical Learning -- 2.1 Classification of Techniques -- 2.2 Supervised Learning -- 2.2.1 Classification -- 2.2.2 Regression -- 2.2.3 Validation of Results: k-Fold Cross-Validation -- 2.3 Unsupervised Learning -- 2.3.1 Clustering -- 2.3.2 Association Rules -- 2.3.3 Expectation-Maximization Algorithm -- 3 On the Classification of Protection Procedures -- 3.1 Dimensions -- 3.1.1 On Whose Privacy Is Being Sought -- 3.1.2 On the Computations to be Done -- 3.1.3 On the Number of Data Sources -- 3.1.4 Knowledge Intensive Data Privacy -- 3.1.5 Other Dimensions and Discussion -- 3.1.6 Summary -- 3.2 Respondent and Holder Privacy -- 3.3 Data-Driven Methods -- 3.4 Computation-Driven Methods -- 3.4.1 Single Database: Differential Privacy -- 3.4.2 Multiple Databases: Cryptographic Approaches -- 3.4.3 Discussion -- 3.5 Result-Driven Approaches -- 3.6 Tabular Data -- 3.6.1 Cell Suppression -- 3.6.2 Controlled Tabular Adjustment -- 4 User's Privacy -- 4.1 User Privacy in Communications -- 4.1.1 Protecting the Identity of the User -- 4.1.2 Protecting the Data of the User -- 4.2 User Privacy in Information Retrieval -- 4.2.1 Protecting the Identity of the User -- 4.2.2 Protecting the Query of the User -- 4.3 Private Information Retrieval -- 4.3.1 Information-Theoretic PIR with k Databases -- 4.3.2 Computational PIR -- 4.3.3 Other Contexts -- 5 Privacy Models and Disclosure Risk Measures -- 5.1 Definition and Controversies