A Privacy-Preserving Similarity Search Scheme over Encrypted Word Embeddings
A Privacy-Preserving Similarity Search Scheme over Encrypted Word Embeddings
复制标题
DOI：
10.1145/3366030.3366081
复制
发表时间：
2019-12
期刊：
影响因子：
--
通讯作者：
Daisuke Aritomo;Chiemi Watanabe;Masaki Matsubara;Atsuyuki Morishima
中科院分区：
文献类型：
--
作者：
Daisuke Aritomo;Chiemi Watanabe;Masaki Matsubara;Atsuyuki Morishima
云计算平台的最新发展吸引了前所未有的大量数据。如今，即使是最敏感的数据也被外包，因此，保护对于确保隐私不被云平台提供的便利所交易至关重要。传统的对称加密方案提供了良好的保护；然而，它们破坏了云计算的优点。已经尝试获得一种可以实现功能性和保护性的方案。然而，现有可搜索加密方案中提供的功能往往落后于信息检索 (IR) 领域的最新发现。在本研究中，我们提出了一种基于 Simhash 的隐私保护相似文档搜索系统。我们的方案对最新的基于机器学习的 IR 方案开放，并且利用基于 VP 树的索引对性能进行了调整，该索引针对安全性进行了优化。对真实数据集的分析和各种测试证明了该方案在真实数据集上的安全性和效率。
Recent evolution in cloud computing platforms have attracted the largest amount of data than ever before. Today, even the most sensitive data are being outsourced, thus, protection is essential to ensure that privacy is not traded for the convenience provided by cloud platforms. Traditional symmetric encryption schemes provide good protection; however, they ruin the merits of cloud computing. Attempts have been made to obtain a scheme where both functionality and protection can be achieved. However, features provided in existing searchable encryption schemes tend to be left behind the latest findings in the information retrieval (IR) area. In this study, we propose a privacy-preserving similar document search system based on Simhash. Our scheme is open to the latest machine-learning based IR schemes, and performance has been tuned utilizing a VP-tree based index, which is optimized for security. Analysis and various tests on real-world datasets demonstrate the scheme's security and efficiency on real-world datasets.