Entropy-Based Correlation Analysis for Privacy Risk Assessment in IoT Identity Ecosystem

Area of Science:

• Cybersecurity
• Information Science
• Computer Science

Background:

• The expanding Internet of Things (IoT) necessitates robust privacy risk assessment tools.
• Existing methods for evaluating IoT privacy vulnerabilities require enhancement.
• Quantitative frameworks are crucial for understanding and mitigating privacy risks in interconnected devices.

Purpose of the Study:

• To introduce a quantitative framework for evaluating IoT privacy risks.
• To analyze the correlation between the Personalized Privacy Assistant (PPA) and PrivacyCheck scores.
• To assess the effectiveness of these scores in detecting privacy vulnerabilities across various sensitive data types.

Main Methods:

• Development of a quantitative framework for IoT privacy risk evaluation.
• Utilizing Bayesian networks with cycle decomposition to model risk factor dependencies.
• Application of entropy-based metrics for quantifying informational uncertainty in privacy assessments.
• Analysis of score correlations across sensitive data types (email, SSN, location).

Main Results:

• The study highlights the strengths and limitations of both the PPA and PrivacyCheck tools.
• Experimental results demonstrate the effectiveness of the proposed framework in identifying privacy vulnerabilities.
• A significant correlation was observed between the two privacy scores across different data types.
• The framework provides a data-driven approach to privacy risk scoring.

Conclusions:

• Combining data-driven risk scoring, information-theoretic analysis, and network modeling offers a comprehensive approach to IoT privacy evaluation.
• The proposed framework enhances the ability to detect and manage privacy risks in IoT environments.
• Further research can refine these metrics for more granular privacy assessments.