Secure Computation of Gaussian Process Regression for Data Analysis
Takayuki Nakachi, Yitu Wang · 2021 29th European Signal Processing Conference (EUSIPCO) · 2021
We propose Gaussian process regression (GPR) for encrypted data generated based on random unitary transformation. Edge cloud computing is spreading into many application fields, including data analysis services. However, new privacy concerns are being raised regarding things such as data leakage from unreliable providers and accidents. Therefore, we examine a secure GPR method that incorporates privacy protection. We prove that our GPR for encrypted data performs the same as GPR for non-encrypted data. Finally, we determine the effectiveness of our method by experimenting with diabetes data from the medical analysis field and synthetic data.