Efficient privacy preservation of big data for accurate data mining

Mahawaga Arachchige Pathum Chamikara, Peter Bertók, Dongxi Liu, Seyit Ahmet Çamtepe, Ibrahim Khalil

Abstract: Highlights•Data mining, accompanied with information dissemination can lead to a privacy breach.•Utility and privacy often appear as conflicting factors in the existing privacy-preserving approaches.•An efficient perturbation paradigm is proposed to provide enough balance between privacy and utility.•A new privacy model is proposed to guarantee optimal perturbation parameter selection.•The proposed data perturbation approach excels in speed, scalability, and accuracy for big data perturbation.