Journal - Volume I No.1 September 2014 , Special Issue-1

  • » Back to Index

  • Title

    :

    A ROBUST DATA OBFUSCATION APPROACH FOR PRIVACY PRESERVING DATAMINING

    Authors

    :

    S.Deebika1 A.Sathyapriya2

    Keywords

    :

    Anonymization, L-Diversity, PPDM, PSensitive, T-Closeness, (n,t)-Closeness

    Issue Date

    :

    September - 2014

    Abstract

    :

    Data mining play an important role in the storing and retrieving of huge data from database. Every user wants to efficiently retrieve some of the encrypted files containing specific keywords, keeping the keywords themselves secret and not jeopardizing the security of the remotely stored files. For well-defined security requirements and the global distribution of the attributes needs the privacy preserving data mining (PPDM). Privacy-preserving data mining is used to uphold sensitive information from unendorsed disclosure. Privacy preserving data is to develop methods without increasing the risk of misuse of the data. Anonymization techniques: K- Anonymity, L-Diversity, T-Closeness, P-Sensitive and M-invariance offers more privacy options rather to other privacy preservation techniques (Randomization, Encryption, and Sanitization). All these Anonymization techniques only offer resistance against prominent attacks like homogeneity and background. None of them is able to provide a protection against all known possible attacks and calculate overall proportion of the data by comparing the sensitive data. We will try to evaluate a new technique called (n,t)-Closeness which requires that the distribution of a sensitive attribute in any equivalence class to be close to the distribution of the attribute in the overall table.

    Page(s)

    :

    16 - 21

    ISSN

    :

    2347- 4734

    Source

    :

    Vol. 1, No.1 - Special Issue-1

    Download

    :


  • » Back index