Description

Electronic Health Records (EHRs) has been used in many healthcare centres in order to improve the patient’s privacy and it would increase the productivity. In order perverse the patient’s specific data, privacy preserving integration method based on ensemble strategy have been used earlier. Each healthcare centre will contribute their data based on the user specific needs. The data is collected about the patients from various healthcare centres. The horizontally partitioned distributed privacy preservation is used, so thatthe data which is collected will be the different subset of records with same set of attributes. While integrating these large data sets, there will be some unwanted models. To prevent this negative impact, there will be one person to decide which of the local models are needed and that will be integrated with it. We are going to extend it with the process of k-anonymity, which includes generalization and suppression techniques. Here the data that is distributed around can be generalized using MinGen Algorithm and the patient’s specific data can be preserved using suppression technique. For this process a healthcare centre needs to share the user specific data so that the patient’s data cannot be determined. Generalization is the process replacing a value with a less specific but consistent value. So the subsets of records are generalized. The MinGen algorithm combines these techniques to provide k-anonymity protection with minimal distortion. Quasi-Identifier is used to collect the background information like demographics (such as Date of Birth, gender)and location details. The data can be suppressed using K-anonymity of Classification Tress Using Suppression (KACTUS) can be used. This technique is used to identify the data that have less influence and are being suppressed. So this suppression leads to the accuracy of information. This leads to precise measurement of data distribution without compromising on patient privacy.