Volume 16 Issue 3 - May 2020

  • 1. Summary statistics of sensitive data using differential privacy

    Authors : Jaseem C K

    Pages : 5-12

    DOI : http://dx.doi.org/10.21172/1.163.02

    Keywords : Differential PrivacySummary StatisticsReconstruction AttackData Analysis

    Abstract :

    The privacy of the ever-growing data in the modern world is a rising concern. This data has the potential to inform many useful insights while it is at stake because of the privacy issue. Recent decades have witnessed a number of cases in which the data were either stolen or personal information was identified from the statistical data. Solving this problem of privacy-preserving data analysis might encourage the flow of more data to be used. Differential Privacy is considered to be one of the state-of-the-art concepts that can help us achieve this goal. It is a strong, mathematical definition of privacy in the context of statistical and machine learning analysis. The project focuses on extracting the summary statistics of student alcohol consumption using differential privacy. The IBM differential privacy library is used to implement this and gain insights. The output is the summary statistics with a conclusive inference.

    Citing this Journal Article :

    Jaseem C K, "Summary statistics of sensitive data using differential privacy", Volume 16 Issue 3 - May 2020, 5-12