Incorporating Functional Encryption (FE) in Privacy Preserving Data Mining (PPDM)

Abstract

Privacy preserving data mining has emerged due to large usage of data in organiza- tions for extracting knowledge from data. Big data uses centralized as well as distributed data and mines knowledge. Privacy preservation of data has become critical asset due to malicious users and society issues. This paper expresses issues in privacy preserving data mining which includes both cryptographic and non-cryptographic approaches. Due to se- curity concern, cryptographic approaches like Homomorphic encryption, Shamir's secret sharing schemes and oblivious transfers are more focused. Usage of these approaches in- creases communication and computation cost of data mining operations obviously. This paper has incorporated new approach in privacy preservation, Functional Encryption (FE). FE uses personalized randomness, bi linear groups for cryptographic key mapping, permutations etc. which makes it more complex, but more e cient. Two algorithms are proposed with Trusted Third Party and Collaborative processing model incorporating FE schemes. FE provides higher level of security and data privacy. FE only allows to learn the output of function without revealing anything else. Final model exhibits feasi- ble computation cost. Communication cost in (Semi Trusted Authority) STA model is O(n3), while in (Semi Trusted Third Party) STTP model, it is O(n2logn).