Toward Practical Privacy-Preserving Frequent Itemset Mining on Encrypted Cloud Data
The main aim of this project is to build a new framework for enforcing privacy in frequent itemset mining on encrypted cloud data.
In our proposed system we propose privacy-preserving framework for secure frequent itemset mining on encrypted data, where only one aided server (referred to as Evaluator) is needed besides the Cloud Service Provider (CSP). The CSP collects encrypted transactions and maintains an encrypted transaction database, and the Evaluator to compute frequent itemset mining efficiently and correctly, and also preserve the privacy of transactions and the mining result. The Evaluator first generates a pair of public/private key. Then, each user encrypts its transactions with the public key of the Evaluator and contributes its encrypted transactions to the CSP, because in our model, users do not fully trust the CSP with the privacy of their transactions. Therefore, all their transactions are encrypted before contributing to the CSP.