Paper Title
A Greedy Generalized Heuristic Approach to Protect Identity Disclosure
Abstract
The entire process of reconstructing original values from anonymized values could be turned away using a
number of random values varying from 1 to 4 levels. This is often called as negative understanding. The negative
understanding implementation prior systems is really a hypothesis and doesn't offer any evidential truth on its influence over
k(m,n)-anonymization procedure. To sustain the efficiency of k(m,n) anonymization procedure, we attempt to demonstrate
the hypothesis using real-time implementation. The paper defines k(mn)-anonymity, which supplies protection against
identity disclosure and proposes a greedy anonymization heuristic that has the capacity to sanitize large datasets. The
formula and the caliber of the anonymization are evaluated experimentally. Collections of real-world data will often have
implicit or explicit structural relations. Within this work, we concentrate on tree structured data. Such data originate from
various programs, even if your structure isn't directly reflected within the syntax, e.g. XML documents. An attribute situation
is really a database where details about an individual is scattered among different tables which are connected through foreign
keys.
Keywords— Privacy, Tree Data, Anonymity, Structural Knowledge, Generalization, Disassociation