A Heuristic Approach to Preserve Privacy in Stream Data with Classification

Data stream Mining is new era in data mining field. Numerous algorithms are used to extract knowledge and classify stream data. Data stream mining gives birth to a problem threat of data privacy. Traditional algorithms are not appropriate for stream data due to large scale. To build classification model for large scale also required some time constraints which is not fulfilled by traditional algorithms. In this Paper we propose a Heuristic approach to preserve privacy with classification for stream data. This approach preserves privacy and also improves process to extract knowledge and build classification model for stream data. This method is implemented in two phases. First is processing of data and second classification analysis. In these two phases first data stream perturbation is applied on data set and after that classification is applied on perturbed data as well as original dataset. Experimental results and charts show that this approach not only preserve privacy but it can also reduces complexity to mine large scale stream data.