期刊名称：International Journal of Computer Science and Network
印刷版ISSN：2277-5420
出版年度：2016
卷号：5
期号：2
页码：381-387
出版社：IJCSN publisher
摘要：Data stream is a continuous and changingsequence of data that continuously arrive at a system to storeor process. It is vital to find out useful information from largeenormous amount of data streams generated from differentapplications viz. organization record, call center record,sensor data, network traffic, web searches etc. Privacypreserving data mining techniques allow generation of datafor mining and preserve the private information of theindividuals. In this paper, classification algorithms wereapplied on original data set as well as privacy preserved dataset. Results were compared to evaluate the performance ofvarious classification algorithms on the data streams that hadbeen privacy preserved using anonymization techniques. Thepaper proposes an effective approach for classification ofanonymized data streams. Intensive experiments wereperformed using appropriate data mining and anonymizationtools. Experimental result shows that the proposed approachimproves accuracy of classification and increases the utility,i.e. accuracy of classification while minimizing the meanabsolute error. The proposed work presents theanonymization technique effective in terms of informationloss and the classifiers efficient in terms of response timeanddata usability.
关键词：Data Mining; Privacy Preservation; Data Stream;Privacy Preservation Data Mining; Anonymization;Classification; ARX-Tool
