On Using Linear Diophantine Equations to Tune the extent of Look Ahead while Hiding Decision Tree Rules
This work addresses privacy concerns for binary datasets by enabling public data use without heuristic restrictions, though it appears incremental as it builds on existing hiding methodologies.
The paper tackles the problem of preserving privacy of sensitive patterns in decision trees by proposing a record augmentation approach that uses linear Diophantine equations to add instances while minimizing disturbance to node entropy, resulting in a method that hides rules with minimal data alteration.
This paper focuses on preserving the privacy of sensitive pat-terns when inducing decision trees. We adopt a record aug-mentation approach for hiding sensitive classification rules in binary datasets. Such a hiding methodology is preferred over other heuristic solutions like output perturbation or crypto-graphic techniques - which restrict the usability of the data - since the raw data itself is readily available for public use. In this paper, we propose a look ahead approach using linear Diophantine equations in order to add the appropriate number of instances while minimally disturbing the initial entropy of the nodes.