dK-Personalization: Publishing Network Statistics with Personalized Differential Privacy

Masooma Iftikhar*, Qing Wang, Yang Li

*Corresponding author for this work

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    5 Citations (Scopus)

    Abstract

    Preserving privacy of an individual in network structured data while enhancing utility of published data is one of the most challenging problems in data privacy. Moreover, different individuals might have different privacy levels based on their own preferences, thereby personalization needs to be considered to achieve personal data protection. In this paper, we aim to develop a privacy-preserving mechanism to publish network statistics, particularly degree distribution, and joint degree distribution, which guarantees personalized (edge or node) differential privacy while enhancing network data utility. To this extend we propose four approaches to handle personal privacy requirements of individuals in a differentially private computation. We have empirically verified the utility enhancement and privacy guarantee of our proposed approaches on four real-world network datasets. To the best of our knowledge, this is the first study to publish network data distributions under personalized differential privacy, while enhancing network data utility.

    Original languageEnglish
    Title of host publicationAdvances in Knowledge Discovery and Data Mining - 26th Pacific-Asia Conference, PAKDD 2022, Proceedings
    EditorsJoão Gama, Tianrui Li, Yang Yu, Enhong Chen, Yu Zheng, Fei Teng
    PublisherSpringer Science and Business Media Deutschland GmbH
    Pages194-207
    Number of pages14
    ISBN (Print)9783031059322
    DOIs
    Publication statusPublished - 2022
    Event26th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2022 - Chengdu, China
    Duration: 16 May 202219 May 2022

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume13280 LNAI
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference26th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2022
    Country/TerritoryChina
    CityChengdu
    Period16/05/2219/05/22

    Fingerprint

    Dive into the research topics of 'dK-Personalization: Publishing Network Statistics with Personalized Differential Privacy'. Together they form a unique fingerprint.

    Cite this