Author : Kumaraswamy S, Manjula S H, K R Venugopal and L M Patnaik

Privacy Preserving Data Mining ( ) presents a novel framework for extracting and deriving information when the data is distributed amongst the multiple parties. The privacy preservation of data and the use of efficient data mining algorithms in systems is a major issue that exists. Most of the existing systems employ the cryptographic key exchange process and the key computation process accomplished by means of certain trusted server or a third party. To eliminate the key exchange and key computation overheads this paper discusses the Key Distribution-Less Privacy Preserving Data Mining ( ) system. The novelty of the system is that no data is published but only the association rules are published to achieve effective data mining results. The embodies the data mining algorithm for classification rule generation and data mining. The results discussed in this paper compare the based system with the based system and the efficiency in rule generation, overhead reduction and classification efficiency of the latter is proved.

Keywords: Privacy preserving data mining, Association rules, C4.5 algorithm, See5.0/C5.0 algorithm, classification rules

Article published in International Journal of Current  Engineering  and Technology, Vol.4,No.2 (April- 2014)