Atjaunināt sīkdatņu piekrišanu

Differential Privacy and Applications 1st ed. 2017 [Hardback]

5.00/5 (1 ratings by Goodreads)
, , ,
  • Formāts: Hardback, 235 pages, height x width: 235x155 mm, weight: 5029 g, 71 Illustrations, black and white; XIII, 235 p. 71 illus., 1 Hardback
  • Sērija : Advances in Information Security 69
  • Izdošanas datums: 08-Sep-2017
  • Izdevniecība: Springer International Publishing AG
  • ISBN-10: 3319620029
  • ISBN-13: 9783319620022
Citas grāmatas par šo tēmu:
  • Hardback
  • Cena: 136,16 €*
  • * ši ir gala cena, t.i., netiek piemērotas nekādas papildus atlaides
  • Standarta cena: 160,19 €
  • Ietaupiet 15%
  • Grāmatu piegādes laiks ir 3-4 nedēļas, ja grāmata ir uz vietas izdevniecības noliktavā. Ja izdevējam nepieciešams publicēt jaunu tirāžu, grāmatas piegāde var aizkavēties.
  • Daudzums:
  • Ielikt grozā
  • Piegādes laiks - 4-6 nedēļas
  • Pievienot vēlmju sarakstam
Differential Privacy and Applications 1st ed. 2017Palielināt
  • Formāts: Hardback, 235 pages, height x width: 235x155 mm, weight: 5029 g, 71 Illustrations, black and white; XIII, 235 p. 71 illus., 1 Hardback
  • Sērija : Advances in Information Security 69
  • Izdošanas datums: 08-Sep-2017
  • Izdevniecība: Springer International Publishing AG
  • ISBN-10: 3319620029
  • ISBN-13: 9783319620022
Citas grāmatas par šo tēmu:
Permanent link: https://www.kriso.lv/db/9783319620022.html
Keywords:
This book focuses on differential privacy and its application with an emphasis on technical and application aspects. This book also presents the most recent research on differential privacy with a theory perspective. It provides an approachable strategy for researchers and engineers to implement differential privacy in real world applications.Early chapters are focused on two major directions, differentially private data publishing and differentially private data analysis. Data publishing focuses on how to modify the original dataset or the queries with the guarantee of differential privacy. Privacy data analysis concentrates on how to modify the data analysis algorithm to satisfy differential privacy, while retaining a high mining accuracy. The authors also introduce several applications in real world applications, including recommender systems and location privacyAdvanced level students in computer science and engineering, as well as researchers and professionals working in

privacy preserving, data mining, machine learning and data analysis will find this book useful as a reference. Engineers in database, network security, social networks and web services will also find this book useful.

Preliminary of Differential Privacy.- Differentially Private Data Publishing: Settings and Mechanisms.- Differentially Private Data Publishing: Interactive Setting.- Differentially Private Data Publishing: Non-interactive Setting.- Differentially Private Data Analysis.- Differentially Private Deep Learning.- Differentially Private Applications: Where to Start .- Differentially Private Social Network Data Publishing.- Differentially Private Recommender System.- Privacy Preserving for Tagging Recommender Systems.- Differential Location Privacy.- Differentially Private Spatial Crowdsourcing.- Correlated Differential Privacy for Non-IID Datasets.- Future Directions.
1 Introduction
1(6)
1.1 Privacy Preserving Data Publishing and Analysis
1(1)
1.2 Privacy Violations
2(1)
1.3 Privacy Models
3(1)
1.4 Differential Privacy
4(1)
1.5 Outline and Book Overview
5(2)
2 Preliminary of Differential Privacy
7(10)
2.1 Notations
7(2)
2.2 Differential Privacy Definition
9(1)
2.2.1 The Privacy Budget
9(1)
2.3 The Sensitivity
10(2)
2.3.1 The Global Sensitivity
11(1)
2.3.2 The Local Sensitivity
11(1)
2.4 The Principle Differential Privacy Mechanisms
12(3)
2.4.1 The Laplace Mechanism
13(1)
2.4.2 The Exponential Mechanism
14(1)
2.5 Utility Measurement of Differential Privacy
15(2)
3 Differentially Private Data Publishing: Settings and Mechanisms
17(6)
3.1 Interactive and Non-interactive Settings
17(2)
3.2 Publishing Mechanism
19(4)
4 Differentially Private Data Publishing: Interactive Setting
23(12)
4.1 Transaction Data Publishing
23(3)
4.1.1 Laplace
23(1)
4.1.2 Transformation
24(1)
4.1.3 Query Separation
24(1)
4.1.4 Iteration
25(1)
4.1.5 Discussion
25(1)
4.2 Histogram Publishing
26(3)
4.2.1 Laplace
27(1)
4.2.2 Partition of Dataset
27(1)
4.2.3 Consistency of Histogram
28(1)
4.3 Stream Data Publishing
29(2)
4.3.1 Laplace
30(1)
4.3.2 Partition of Dataset
30(1)
4.3.3 Iteration
30(1)
4.3.4 Discussion
31(1)
4.4 Graph Data Publishing
31(3)
4.4.1 Edge Differential Privacy
32(1)
4.4.2 Node Differential Privacy
33(1)
4.4.3 Discussion
34(1)
4.5 Summary on Interactive Setting
34(1)
5 Differentially Private Data Publishing: Non-interactive Setting
35(14)
5.1 Batch Queries Publishing
35(4)
5.1.1 Laplace
36(1)
5.1.2 Transformation
36(2)
5.1.3 Partition of Dataset
38(1)
5.1.4 Iteration
38(1)
5.1.5 Discussion
38(1)
5.2 Contingency Table Publishing
39(2)
5.2.1 Laplace
39(1)
5.2.2 Iteration
40(1)
5.2.3 Transformation
40(1)
5.3 Anonymized Dataset Publishing
41(2)
5.4 Synthetic Dataset Publishing
43(5)
5.4.1 Synthetic Dataset Publishing Based on Learning Theory
43(4)
5.4.2 High Dimensional Synthetic Dataset Publishing
47(1)
5.5 Summary on Non-interactive Setting
48(1)
6 Differentially Private Data Analysis
49(18)
6.1 Laplace/Exponential Framework
49(8)
6.1.1 SuLQ and PINQ Interface
50(1)
6.1.2 Specific Algorithms in the Laplace/Exponential Framework
51(6)
6.1.3 Summary on Laplace/Exponential Framework
57(1)
6.2 Private Learning Framework
57(8)
6.2.1 Foundation of ERM
58(1)
6.2.2 Private Learning in ERM
59(3)
6.2.3 Sample Complexity Analysis
62(2)
6.2.4 Summary on Private Learning Framework
64(1)
6.3 Summary of Differentially Private Data Analysis
65(2)
7 Differentially Private Deep Learning
67(16)
7.1 Introduction
67(2)
7.2 Preliminary
69(4)
7.2.1 Deep Learning Structure
69(2)
7.2.2 Stochastic Gradient Descent
71(2)
7.3 Differentially Private Deep Learning
73(8)
7.3.1 Basic Laplace Method
74(1)
7.3.2 Private SGD Method
75(2)
7.3.3 Deep Private Auto-Encoder Method
77(2)
7.3.4 Distributed Private SGD
79(2)
7.4 Experimental Methods
81(1)
7.4.1 Benchmark Datasets
81(1)
7.4.2 Learning Objectives
81(1)
7.4.3 Computing Frameworks
82(1)
7.5 Summary
82(1)
8 Differentially Private Applications: Where to Start?
83(8)
8.1 Solving a Privacy Problem in an Application
83(2)
8.2 Challenges in Differentially Private Applications
85(3)
8.2.1 High Sensitivity Challenge
85(1)
8.2.2 Dataset Sparsity Challenge
85(1)
8.2.3 Large Query Set Challenge
86(1)
8.2.4 Correlated Data Challenge
86(1)
8.2.5 Computational Complexity Challenge
87(1)
8.2.6 Summary
87(1)
8.3 Useful Public Datasets in Applications
88(2)
8.3.1 Recommender System Datasets
88(1)
8.3.2 Online Social Network Datasets
89(1)
8.3.3 Location Based Datasets
89(1)
8.3.4 Other Datasets
89(1)
8.4 Applications Settings
90(1)
9 Differentially Private Social Network Data Publishing
91(16)
9.1 Introduction
91(1)
9.2 Preliminaries
92(1)
9.3 Basic Differentially Private Social Network Data Publishing Methods
93(5)
9.3.1 Node Differential Privacy
93(4)
9.3.2 Edge Differential Privacy
97(1)
9.4 Graph Update Method
98(7)
9.4.1 Overview of Graph Update
98(2)
9.4.2 Graph Update Method
100(1)
9.4.3 Update Function
101(1)
9.4.4 Privacy and Utility Analysis
101(2)
9.4.5 Experimental Evaluation
103(2)
9.5 Summary
105(2)
10 Differentially Private Recommender System
107(24)
10.1 Introduction
107(2)
10.2 Preliminaries
109(3)
10.2.1 Collaborative Filtering
109(1)
10.2.2 Neighborhood-Based Methods: κ Nearest Neighbors
109(2)
10.2.3 Model-Based Methods: Matrix Factorization
111(1)
10.3 Basic Differentially Private Recommender Systems
112(5)
10.3.1 Differentially Private Untrustworthy Recommender System
113(1)
10.3.2 Differentially Private Trustworthy Recommender System
114(3)
10.4 Private Neighborhood-Based Collaborative Filtering Method
117(12)
10.4.1 KNN Attack to Collaborative Filtering
117(1)
10.4.2 The Private Neighbor Collaborative Filtering Algorithm
118(5)
10.4.3 Privacy and Utility Analysis
123(3)
10.4.4 Experiment Analysis
126(3)
10.5 Summary
129(2)
11 Privacy Preserving for Tagging Recommender Systems
131(20)
11.1 Introduction
131(2)
11.2 Preliminaries
133(1)
11.2.1 Notations
133(1)
11.2.2 Tagging Recommender Systems
133(1)
11.2.3 Related Work
134(1)
11.3 Private Tagging Publishing Method
134(15)
11.3.1 User Profiles
134(2)
11.3.2 Private Tagging Release Algorithm Overview
136(1)
11.3.3 Private Topic Model Generation
137(2)
11.3.4 Topic Weight Perturbation
139(2)
11.3.5 Private Tag Selection
141(2)
11.3.6 Privacy and Utility Analysis
143(3)
11.3.7 Experimental Evaluation
146(3)
11.4 Summary
149(2)
12 Differentially Location Privacy
151(22)
12.1 Introduction
151(1)
12.2 Preliminary
152(1)
12.3 Basic Location Privacy Methods
153(7)
12.3.1 Snapshot Location Privacy: Geo-Indistinguishability
154(3)
12.3.2 Trajectory Privacy
157(3)
12.4 Hierarchical Snapshot Location Publishing
160(12)
12.4.1 Hierarchical Sensitivity
160(2)
12.4.2 Overview of Private Location Release
162(1)
12.4.3 Private Location Release Algorithm
163(4)
12.4.4 Utility and Privacy
167(3)
12.4.5 Experimental Evaluation
170(2)
12.5 Summary
172(1)
13 Differentially Private Spatial Crowdsourcing
173(18)
13.1 Introduction
173(1)
13.2 Basic Method
174(3)
13.2.1 Background of Crowdsourcing
174(1)
13.2.2 Differentially Private Crowdsourcing Methods
175(2)
13.3 Differential Privacy in Reward-Based Crowdsourcing
177(12)
13.3.1 Problem Statement
178(1)
13.3.2 Building a Contour Plot with DP Guarantee
178(3)
13.3.3 Task Assignment
181(5)
13.3.4 Experimental Evaluation
186(3)
13.4 Summary
189(2)
14 Correlated Differential Privacy for Non-IID Datasets
191(24)
14.1 Introduction
191(1)
14.2 An Example: Correlated Records in a Dataset
192(2)
14.3 Basic Methods
194(2)
14.3.1 Pufferfish
194(1)
14.3.2 Blowfish
195(1)
14.4 Correlated Differential Privacy
196(18)
14.4.1 Correlated Differential Privacy Problem
196(1)
14.4.2 Research Issues and Challenges
197(1)
14.4.3 Correlated Dataset Analysis
198(1)
14.4.4 Correlated Sensitivity
199(2)
14.4.5 Correlated Iteration Mechanism
201(5)
14.4.6 Mechanism Analysis
206(2)
14.4.7 Experiment and Analysis
208(6)
14.5 Summary
214(1)
15 Future Directions and Conclusion
215(8)
15.1 Adaptive Data Analysis: Generalization in Machine Learning
215(1)
15.2 Personalized Privacy
216(1)
15.3 Secure Multiparty Computations with Differential Privacy
216(1)
15.4 Differential Privacy and Mechanism Design
217(1)
15.5 Differential Privacy in Genetic Data
217(1)
15.6 Local Differential Privacy
218(2)
15.7 Learning Model Publishing
220(2)
15.8 Conclusion
222(1)
References 223(12)
Index 235