Privacy-Preserving Outsourcing Scheme for SVM on Vertically Partitioned Data

Support vector machine (SVM) is an important technique for data classification. Traditional SVM assumes free access to data. If the are split and held by different users, privacy reasons, users likely unwilling submit their a third party In this paper, using additive homomorphic encryption random transformations (matrix transformation decomposition), we design privacy-preserving outsourcing scheme conducting Least Squares (LS-SVM) classification on vertically partitioned our system, multiple owners (users) encrypted two non-colluding service providers, which conduct algorithm it. During execution of algorithm, neither provider learns anything about input data, intermediate results, or predicted result. other words, in whole process. Extensive theoretical analysis experimental evaluation demonstrate correctness, security, efficiency method.