This book presents the concept of a fuzzy-based recommender system for user account privacy settings that can be used for citizen participation on online political platforms. The elaborated components are exemplarily based on the needs of a political platform implemented during the presidential election in Ecuador. The book readdresses the issue of privacy paradox demonstrating that, indeed, users’ actual decisions of being private in most cases diverge with their initial privacy intentions. The two concepts presented in the book - the citizen privacy profile framework and the prototype fuzzy-based privacy settings recommender system - can be adapted by different organizations such as government institutions, NGOs, or private online service providers to meet their specific needs.
The book will be of interest to researchers and practitioners in the areas of usage modeling, privacy, system design, and for service providers in eDemocracy.
Author(s): Aigul Kaskina
Series: Fuzzy Management Methods
Publisher: Springer
Year: 2022
Language: English
Pages: 120
City: Cham
Foreword
Acknowledgments
Contents
List of Figures
List of Tables
List of Acronyms
1 Introduction
1.1 Motivation
1.2 Objectives
1.3 Research Questions
1.4 Research Methods
1.5 Thesis Outline
1.6 Own Research Contribution
References
2 Insights into Privacy Research
2.1 Economics of Privacy—a Rational Human
2.1.1 Privacy Calculus
2.1.2 Death to Privacy Calculus?
2.1.3 Privacy Paradox
2.2 Beyond Rationality—Human Cognition
2.2.1 Cognitive Heuristics Behind Disclosure Decisions
2.2.2 Enforced by Individual Characteristics?
2.3 Automating User Privacy Support
References
3 Citizen Privacy Profile Framework
3.1 Conceptual Development
3.1.1 Overview of Existing Privacy Frameworks
Liu et al. framework
Aimeur et al. Framework
Knijnenburg et al. Framework
3.1.2 Citizen Privacy Profile (CPP) Framework
Voting Advice Applications
Citizen Privacy Profile Framework
3.1.3 Evaluation of CPP Framework
3.2 Implementation of CPP
3.2.1 Platform Description
Platform Blocks
Type of Content
User Roles
3.2.2 Privacy Settings Functionality
Information: Data Levels
Social Network: Audience Levels
3.2.3 Privacy Profiles Extraction
3.3 User Privacy Profiles Modelling
3.3.1 Fuzzy Clustering
Dataset Clustering Tendency
Fuzzy C-Means Clustering
Partitioning Around Medoids Clustering
Measuring Distances
3.3.2 Evaluation of Clustering Validity
Partition Coefficient (PC)
Partition Entropy (PE)
Modified Partition Coefficient (MPC)
Xie and Beni Index (XBI)
Crisp Silhouette (CS)
Fuzzy Silhouette (FS)
Analysis of the Validation Results
3.3.3 Discussion of Fuzzy Privacy Profiles
3.4 User Evaluation of Privacy Behaviour
3.4.1 Objective System Aspect
3.4.2 User Experience
3.4.3 Personal Characteristics
3.4.4 Evaluation Analysis
Independent Variables Analysis
Combined Variable Analysis
3.5 Conclusions
References
4 Fuzzy-Based Privacy Settings Recommender System
4.1 Conceptual Development
4.1.1 Overview of the Existing Systems
Privacy Wizard
YourPrivacyProtector
Personalised Privacy Assistant
4.1.2 Fuzzy-Based Privacy Settings Recommender System (FPRS)
4.2 Implementation of FPRS
4.2.1 Prototype Architecture
4.2.2 Recommendations Calculation
4.2.3 Evaluation of Recommendations Accuracy
4.3 User Evaluation of Privacy Recommendations
4.3.1 Objective System Aspects
4.3.2 Subjective System Aspects
Perceived Recommendation Sharpness
Perceived Recommendation Quality
4.3.3 User Experience
Persuasion Effect
Reactance Effect
Outcome Satisfaction
4.3.4 Personal Characteristics
Inconsistent/Consistent Privacy Behaviour
4.3.5 Evaluation Setup
Experimental Manipulations
Experimental Procedures
Defining Measurements
4.3.6 Evaluation Analysis
4.4 Conclusions
References
5 Conclusions
5.1 Discussions
5.1.1 Implication to the Public Need
5.1.2 Implication to the Knowledge Base
5.1.3 Limitations of the Research
5.1.4 Answering Research Questions
5.2 Future Outlook
References
Appendix A
A.1 Privacy Profiles Extraction
Appendix B
B.1 User Evaluation Survey for Privacy Behaviour
Appendix C
C.1 User Evaluation Setup
