Privacy-Preserving Analysis Technique for Secure, Cloud-based Data Mining with Cloud Service Provider

Abstract

These days, data mining is commonly used to find relationships between the elements in large datasets. Frequent itemset mining is an essential component of association rule mining, one of the most widely used techniques for data mining. The truthful but inquisitive cloud service provider (CSP) receives large amounts of data. Encrypting data before uploading it to the cloud is generally acknowledged as a way to protect its privacy. This makes it difficult to analyze data, particularly association rule mining while maintaining data privacy. The smooth integration of information is also threatened by recent developments in data mining and knowledge discovery, which make it possible to uncover buried knowledge in massive amounts of data. We examine the challenge of developing privacy-preserving algorithms for association rule mining, one data mining technique. Technologies that enable privacy-aware outsourcing of sensitive data processing and storage to public clouds are covered in this survey to address this problem. Big data and cloud computing are two recent developments. Therefore, it is crucial to identify the connections between them and extract relevant patterns and knowledge from published publications in these disciplines. Additionally, we provide a list of numerous research initiatives and products that have made some of the ideas surveyed a reality. Lastly, we list unresolved research issues.