Authors: Jing Wang, Libing Wu, Huaqun Wang, Kim-Kwang Raymond Choo, and Debiao He
Title: An Efficient and Privacy-preserving Outsourced Support Vector Machine Training for Internet of Medical Things
Journal: IEEE Internet of Things Journal

Abstract: As the use of machine learning in Internet of Medical Things (IoMT) settings increases, so do the data privacy concerns. Therefore, in this paper we propose an efficient privacy-preserving outsourced support vector machine (SVM) scheme, EPoSVM, designed for IoMT deployment. To securely train the SVM, we design eight secure computation protocols to allow the cloud server to efficiently execute basic integer and floating-point computations. The proposed scheme protects training data privacy and guarantees the security of the trained SVM model. The security analysis proves that our proposed protocols and EPoSVM satisfy both security and privacy protection requirements. Findings from the performance evaluation using two real-world disease datasets also demonstrate the efficiency and effectiveness of EPoSVM in achieving the same classification accuracy as a general SVM.

摘要：机器学习在医疗物联网（IoMT）场景中的应用不断增加。但是，在敏感的医疗保健领域中使用机器学习存在与数据隐私相关的问题。因此，在本文中，我们提出了一种针对IoMT环境的高效隐私保护外包支持向量 机（SVM）方案，称为EPoSVM。为了安全地训练SVM，我们设计了八种安全的隐私计算子协议，以允许云服务器有效地执行基本的整数运算和浮点运算。我们提出的方案保护了训练数据的机密性，并保证了最终SVM模型的安全性。安全性分析证明，我们提出的子协议和EPoSVM可以同时满足安全性和隐私保护要求。使用两个实际疾病数据集进行性能评估的结果也证明了EPoSVM与普通SVM训练上具有相同的分类精度方面和有效性。