Privacy-preserving data collection for 1: M dataset

Generation of humongous data in recent times is due to the ever-growing sources, primarily the Internet of Things, which is then collected to develop effective Artificial Intelligence (AI) and machine learning (ML) solutions. Though such data provides useful insight on various trends that eventually results in better quality of life, collecting such data raises privacy concerns for data owners. Preserving the privacy of individuals during the process of data collection is an important problem specifically in the context of 1:M datasets (an individual can have multiple records). Therefore, a novel privacy-preserving data collection protocol, for 1: M datasets has been proposed in this paper. The privacy-preserving mechanism ensures data safety from external and internal privacy breaches and its effective usage in micro data analysis through AI and ML methods. The use of the leader election algorithm and the notion of l-anatomy minimize the risk of privacy disclosures and enabled us to achieve higher computational efficiency.

Automated summary

A novel privacy-preserving data collection protocol has been proposed in this paper for 1:M datasets, which ensures data safety, minimizes the risk of privacy disclosures, and achieves higher computational efficiency through the use of AI and ML methods.