Comparison of machine learning models applied on anonymized data with different techniques
This work addresses the trade-off between privacy and utility in anonymized datasets for classification tasks, but it is incremental as it applies existing methods to a known dataset.
The study compared four classical machine learning models on anonymized data using techniques like k-anonymity, l-diversity, t-closeness, and δ-disclosure privacy, finding that performance varied with anonymization parameters and techniques applied.
Anonymization techniques based on obfuscating the quasi-identifiers by means of value generalization hierarchies are widely used to achieve preset levels of privacy. To prevent different types of attacks against database privacy it is necessary to apply several anonymization techniques beyond the classical k-anonymity or $\ell$-diversity. However, the application of these methods is directly connected to a reduction of their utility in prediction and decision making tasks. In this work we study four classical machine learning methods currently used for classification purposes in order to analyze the results as a function of the anonymization techniques applied and the parameters selected for each of them. The performance of these models is studied when varying the value of k for k-anonymity and additional tools such as $\ell$-diversity, t-closeness and $δ$-disclosure privacy are also deployed on the well-known adult dataset.