Analyzing the Fitting Effect of Approximate Functions in Homomorphic Encryption Process Under Cloud Computing

In order to ensure the security of machine learning data in the process of cloud computing, we use Taylor's theorem and least square method to approximate the Sigmoid function involved in the privacy protection logistic regression algorithm, and then analyze the fitting effect of approximate fu...

Full description

Saved in:
Bibliographic Details
Published in2022 4th International Conference on Applied Machine Learning (ICAML) pp. 1 - 4
Main Authors Shang, Jingtao, Xiao, Yuqing, Luo, Yin, Liu, Zhijia, Liu, Yunli
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2022
Subjects
Approximation algorithms
Cloud computing
Data privacy
Fitting
Least Square Method
Logistic Regression
Machine learning
Machine learning algorithms
Security
Taylor's Theorem
Online AccessGet full text

Cover

More Information
Summary:In order to ensure the security of machine learning data in the process of cloud computing, we use Taylor's theorem and least square method to approximate the Sigmoid function involved in the privacy protection logistic regression algorithm, and then analyze the fitting effect of approximate functions. The results show that he Taylor theorem is better than the least square method in fitting sigmoid function.
DOI:10.1109/ICAML57167.2022.00061