Logistic Regression and Data Analysis on Privacy Methods on Data Streams
P. Chandrakanth, Anbarasi M.S · International Journal of Engineering & Technology · 2018
The problem data privacy in streams is completely put in a myopic view by hitherto researchers. Research and experimentations have been well fortified on static data, in which predominantly spelled easy with approaches based on perturbation using random data values. Approaches based on large data sets and high dimension data sets are not adequate consequences. By using the phenomenon of autocorrelation of multivariable streams and their leveraging structures, identifying the suitable areas to add noise maximally preserves privacy and in a irreversible manner. Drift checking and ensemble classifier building is the basic requirements for privacy preserving data stream, which makes clear in experimentation with the support of sensitivity analysis. In this paper we present the results of experimentation at all the stages.