Frequent Itemset Mining from Outsourced Transactional Data with Assured Privacy over Cloud Storage

Abstract

Cloud computing is a form of computing in which a large number of remote servers are managed to allow for centralized data storage and online communication between different resources, while Privacy-preserving data mining (PPDM) is a recent trend in privacy and security research. It is defined by the right to privacy, which is one of the most important positioning problems of the information age. An enterprise that lacks computational resources can outsource its mining needs to an outside service provider by using cloud computing services. The elements and association rules of the deployed database, on the other hand, are considered to be the organization's private property. The data owner converts and sends data to the server, ships mining queries to the server and retrieves the actual design for the external server to avoid corporate privacy. In this concept, we examine the issue of outsourcing structures to govern the association within a context that protects company confidentiality. The Rob Frugal method was developed to circumvent the security responsibilities of outsourced data. This approach is an encryption scheme based on one-to-one substitution ciphers for objects and a bogus pattern from the database. In this scheme, the attacker discovers data through a guessing attack, as well as a man in the middle attack, which is possible with Rob Frugal encryption. To overcome this issue, the proposed technique includes Paillier encryption for evaluating the performance level for outsourced data with less complexity and to protect against forging the contents of correspondence. The FP-growth algorithm is used to generate association rules to improve efficiency, and the Paillier cryptosystem is used to maintain a homomorphic encryption algorithm.