.png){kind=icon}
.png){kind=icon}
Return
Quantifying the privacy strength of negative survey based on information entropy and channel capacity
.png){kind=icon}
PRE
AI
刘
W
L
K
DOI：10.1080/27684830.2025.2494917.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Negative surveys protect participant data privacy by collecting only negative information, avoiding direct collection of sensitive personal data and enabling decentralized data reconstruction. Previous research methods relied on simplified assumptions about the original data distribution, assuming that the trade-off between privacy and accuracy is fixed. This study proposes a novel approach using information entropy and channel capacity to quantify privacy strength and accuracy, allowing for more realistic handling of original distributions. We treat the negative survey as an interference channel, measuring accuracy by average mutual information between negative and positive categories, and privacy strength by loss entropy of the positive category. The trade-off between privacy strength and accuracy depends on the original distribution but remains acceptable. Considering initial distribution variability, it's plausible to enhance both accuracy and privacy strength simultaneously. Simulation experiments demonstrate the feasibility of this approach in quantifying negative survey accuracy and privacy strength.
Keywords:
Privacy protection
privacy strength
immune computation
negative survey
information entropy
channel capacity
Journal
R
IF:
1.1
Papers:
72
Citations:
0
Organization
Cited Papers