Collaborative classification mechanism for privacy-Preserving on horizontally partitioned data
We propose a novel two-party privacy-preserving classification solution called Collaborative Classification Mechanism for Privacy-preserving ( $ {\rm C}^{2}{\rm MP}^{2} $ ) over horizontally partitioned data that is inspired from the fact, that global and local learning can be independently executed...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2019-01-01
|
| Series: | Automatika |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/00051144.2019.1578039 |
| Summary: | We propose a novel two-party privacy-preserving classification solution called Collaborative Classification Mechanism for Privacy-preserving ( $ {\rm C}^{2}{\rm MP}^{2} $ ) over horizontally partitioned data that is inspired from the fact, that global and local learning can be independently executed in two parties. This model collaboratively trains the decision boundary from two hyper-planes individually constructed by its own privacy data and global data. $ {\rm C}^{2}{\rm MP}^{2} $ can hide true data entries and ensure the two-parties' privacy. We describe its definition and provide an algorithm to predict future data point based on Goethals's Private Scalar Product Protocol. Moreover, we show that $ {\rm C}^{2}{\rm MP}^{2} $ can be transformed into existing Minimax Probability Machine (MPM), Support Vector Machine (SVM) and Maxi–Min Margin Machine ( $ {\rm M}^4 $ ) model when privacy data satisfy certain conditions. We also extend $ {\rm C}^{2}{\rm MP}^{2} $ to a nonlinear classifier by exploiting kernel trick. Furthermore, we perform a series of evaluations on real-world benchmark data sets. Comparison with SVM from the point of protecting privacy demonstrates the advantages of our new model. |
|---|---|
| ISSN: | 0005-1144 1848-3380 |