Privacy-Preserved Multi-Party Data Merging with Secure Equality Evaluation
Shu Qin Ren, Tan Hong Meng, Yibin Ng, Khin Mi Mi Aung · 2016
Large volumes of customers' information is routinely collected and stored on various of e-service platforms. Data analysis on these data could help decipher the information that truly counts, and it further uncovers hidden patterns, unknown correlations and comprehensive overview for better decision making. However, customers' personal privacy breaches occur in this process despite regulations on the collection, storage and trading of personal data. There is a strong demand for personal privacy protection enforcement starting from data collection, merging to analysis. This paper is mainly address the privacy protection for collection and merging. Its main objective is for secure multi-party data sharing without trusted third-party involvement. Such a merging result will not violate individual privacy yet enable further analysis can be executed on the big data set. For example, medical records from di erent clinics can be securely stored on third-party service center for comprehensive analysis without leaking individual patient's privacy. The main contribution of this work includes: 1) A searchable encryption to provide secure equality test, 2) Independent and exible privacy protection at each party's site, 3) Secure data merging on the collected multi-party data. Our experiments show that merging of 100,000 entries of data takes about 3 minutes while the encryption time is less than 1 seconds. We did the experiment with single process at Intel Core 2 Duo CPU @2.33GHz and 4GB RAM.The results show the encryption and merging schemes can be practically used to operate secure data merging and analysis on untrusted en- vironment, such as cloud computing environment.