Privacy Measurement for Social Network Actor Model
Authors:
- Yong WangYWYong Wang
- Raj Kumar NepaliRNRaj Kumar Nepali
- 1 September 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 659-664
Abstract
Privacy measurement is critical to evaluate privacy risks for business, public policy and legislation. However, there is a lack of effective and practical way to quantify, measure, and evaluate privacy. In this paper, we propose three privacy indexes for privacy measurement (ranking) for social network actor model, i.e., weighted privacy index (w-PIDX), maximum privacy index (m-PIDX), and composite privacy index (c-PIDX). We also introduce a novel virtual attribute for social network actor model to describe the combined attributes' behavior. We further evaluate and demonstrate the effectiveness of these PIDXes for various user groups in different testing scenarios. Our tests and analysis show that composite privacy index, c-PIDX, is the best to measure privacy for social network actor model. A practical approach to evaluate w-PIDX, m-PIDX, and c-PIDX is also presented in the paper.Keywords
This publication has 15 references indexed in Scilit:
- Privacy-preserving resource evaluation in social networks2012
- Privacy in Social Networks: How Risky is Your Social Graph?2012
- Inferring social ties across heterogenous networks2012
- Analyzing facebook privacy settings2011
- A Framework for Computing the Privacy Scores of Users in Online Social NetworksACM Transactions on Knowledge Discovery From Data, 2010
- Privometer: Privacy protection in social networks2010
- Privacy-enabling social networking over untrusted networks2009
- Preserving Privacy in Social Networks Against Neighborhood Attacks2008
- L-diversity: privacy beyond k-anonymity2006
- Privacy in e-commerce1999