Differentially Private Naive Bayes Classifier using Smooth Sensitivity
This work addresses privacy concerns for users in machine learning applications, but it is incremental as it builds on existing differential privacy techniques for a specific algorithm.
The authors tackled the problem of protecting individual privacy in Naive Bayes classifiers by introducing a differentially private version that adds noise based on Smooth Sensitivity, resulting in significantly improved accuracy while maintaining ε-differential privacy compared to methods using global sensitivity.
With the increasing collection of users' data, protecting individual privacy has gained more interest. Differential Privacy is a strong concept of protecting individuals. Naive Bayes is one of the popular machine learning algorithm, used as a baseline for many tasks. In this work, we have provided a differentially private Naive Bayes classifier that adds noise proportional to the Smooth Sensitivity of its parameters. We have compared our result to Vaidya, Shafiq, Basu, and Hong in which they have scaled the noise to the global sensitivity of the parameters. Our experiment results on the real-world datasets show that the accuracy of our method has improved significantly while still preserving $\varepsilon$-differential privacy.