A perspective on data sharing in digital food safety systems.

Critical reviews in food science and nutrition(2022)

引用 3|浏览29
暂无评分
摘要
In this age of data, digital tools are widely promoted as having tremendous potential for enhancing food safety. However, the potential of these digital tools depends on the availability and quality of data, and a number of obstacles need to be overcome to achieve the goal of digitally enabled "smarter food safety" approaches. One key obstacle is that participants in the food system and in food safety often lack the willingness to share data, due to fears of data abuse, bad publicity, liability, and the need to keep certain data (e.g., human illness data) confidential. As these multifaceted concerns lead to tension between data utility and privacy, the solutions to these challenges need to be multifaceted. This review outlines the data needs in digital food safety systems, exemplified in different data categories and model types, and key concerns associated with sharing of food safety data, including confidentiality and privacy of shared data. To address the data privacy issue a combination of innovative strategies to protect privacy as well as legal protection against data abuse need to be pursued. Existing solutions for maximizing data utility, while not compromising data privacy, are discussed, most notably differential privacy and federated learning.
更多
查看译文
关键词
Food safety,data privacy,data utility,decision support tools,predictive models
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