Secure Crowdsourcing Classification with Privacy Protection against Servers

通过针对服务器的隐私保护来确保众包分类的安全

• 批准号: DP220100003
• 负责人: Prof Willy Susilo
• 金额: $ 33.28万
• 依托单位: University of Wollongong
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2022
• 资助国家: 澳大利亚
• 起止时间: 2022-03-01 至 2025-02-28
• 项目状态: 未结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP220100003
• 关键词: Secure; Crowdsourcing; Classification; Privacy; Protection

This project aims to enable comprehensive quality data classification via secure crowdsourcing. The quality of a data-intensive process, such as a Machine Learning algorithm, depends on the input data quality. By using a crowdsourcing classification, the project expects to overcome the painstaking and costly process of humans correctly annotating extensive input data from diverse real information. The expected outcomes are innovative technologies, guaranteeing accuracy and confidentiality of annotation results whilst protecting the privacy of data classification results. It enhances data-intensive outputs quality, which will benefit large data-intensive applications, such as cybersecurity protections via intrusion detection.

该项目旨在通过安全众包实现全面的质量数据分类。数据密集型过程（例如机器学习算法）的质量取决于输入数据的质量。通过使用众包分类，该项目期望克服人类从不同的真实信息中正确注释大量输入数据的艰苦且昂贵的过程。预期成果是创新技术，保证注释结果的准确性和机密性，同时保护数据分类结果的隐私。它提高了数据密集型输出质量，这将有利于大型数据密集型应用程序，例如通过入侵检测进行网络安全保护。