Privacy-preserving data mashup model for trading person-specific information

Abstract

Business enterprises adopt cloud integration services to improve collaboration with their trading partners and to deliver quality data mining services. Data-as-a-Service (DaaS) mashup allows multiple enterprises to integrate their data upon the demand of consumers. Business enterprises face challenges not only to protect private data over the cloud but also to legally adhere to privacy compliance rules when trading person-specific data. They need an effective privacy-preserving business model to deal with the challenges in emerging markets. We propose a model that allows the collaboration of multiple enterprises for integrating their data and derives the contribution of each data provider by valuating the incorporated cost factors. This model serves as a guide for business decision-making, such as estimating the potential risk and finding the optimal value for publishing mashup data. Experiments on real-life data demonstrate that our approach can identify the optimal value in data mashup for different privacy models, including K-anonymity, LKC-privacy, and ∊-differential privacy, with various anonymization algorithms and privacy parameters.