Privacy-preserving Collaborative Computation: Methods, Challenges and Directions.
2023 International Conference on Computer and Applications (ICCA)(2023)
摘要
Although data mining is very relevant to the medical sector, it has also raised privacy concerns since it is applied to sensitive data, which undoubtedly affects citizens’ rights and freedoms, which are strictly regulated by the EU through the General Data Protection Regulation (GDPR). This concern creates a big gap between the data owner and the data analyst, and it is not easy to connect them. Thus, it is evidently important to ensure privacy. This need for privacy becomes a necessity when data from multiple entities aim to collaborate. To tackle this gap, several techniques worth mentioning can be employed during data analysis to ensure privacy, including secure multiparty computation, homomorphic encryption, and federated learning. In this paper, we present the state-of-the-art of existing approaches and discuss their drawbacks to finally identify outstanding challenges in this field.
更多查看译文
关键词
privacy-preserving,secure multiparty computation,homomorphic encryption,federated learning
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要