On the Design and Analysis of the Privacy-Preserving SVM Classifier

Abstract

The support vector machine (SVM) is a widely used tool in classification problems. The SVM trains a classifier by solving an optimization problem to decide which instances of the training data set are support vectors, which are the necessarily informative instances to form the SVM classifier. Since support vectors are intact tuples taken from the training data set, releasing the SVM classifier for public use or shipping the SVM classifier to clients will disclose the private content of support vectors. This violates the privacy-preserving requirements for some legal or commercial reasons. The problem is that the classifier learned by the SVM inherently violates the privacy. This privacy violation problem will restrict the applicability of the SVM. To the best of our knowledge, there has not been work extending the notion of privacy preservation to tackle this inherent privacy violation problem of the SVM classifier. In this paper, we exploit this privacy violation problem, and propose an approach to postprocess the SVM classifier to transform it to a privacy-preserving classifier which does not disclose the private content of support vectors. The postprocessed SVM classifier without exposing the private content of training data is called Privacy-Preserving SVM Classifier (abbreviated as PPSVC). The PPSVC is designed for the commonly used Gaussian kernel function. It precisely approximates the decision function of the Gaussian kernel SVM classifier without exposing the sensitive attribute values possessed by support vectors. By applying the PPSVC, the SVM classifier is able to be publicly released while preserving privacy. We prove that the PPSVC is robust against adversarial attacks. The experiments on real data sets show that the classification accuracy of the PPSVC is comparable to the original SVM classifier.

Wavelet Packet Sub-band Based Classification of Alcoholic and Controlled State EEG Signals

“…For accurate classification only support vectors are required while rest of the dataset becomes redundant. Mathematically, SVM can be considered as ( , ) [3]. Hence, the subjective function ( ) can be written as, (6) where, is the weight matrix.…”

“…Hence, the subjective function ( ) can be written as, (6) where, is the weight matrix. SVM optimizes above equation under the objective function [3], (7) Above equation generates coefficients of hyper-plane which maximizes separation between and . This places features vectors corresponding positive class on one of the sides of the hyperplane, while feature vectors corresponding to negative class resides on the opposite side of the hyper-plane.…”

“…If basic wavelet function has a finite duration, frequency information obtained from WT seems to be localized in time. Therefore, for non-stationary transient signals such as EEG, WT is superior to FFT as well as STFT [3]. In this paper, we focus on special class of WT, the Wavelet Packet Transform to extract subband based features of EEG signals.…”

Privacy-Preserving Clinical Decision Support System Using Gaussian Kernel-Based Classification

“…The recent work in [37] discusses the issue of releasing the trained SVM classifier without violating the privacy of support vectors. While the Gaussian kernel was considered, a Taylor series was exploited to approximate the infinite dimension of the Gaussian kernel into finite dimension and adhere negligible performance loss.…”

An Intelligent Dynamic MRI System for Automatic Nasal Tumor Detection

“…Zhou et al [8] use knowledge-based fuzzy clustering (KBFC), maximum likelihood, and seed growing to identify tumor region and analyze each method's accuracy. Support Vector Machines (SVMs) [9,10] have been widely used in tumor segmentation. Zhang et al [11] compare the tumor segmentation results with one-class SVM and two-class SVM.…”