Basic Informations

C.V

database
web development
project management
system analysis and design

Master Title

Securing the Privacy of Healthcare Data During Publishing

Master Abstract

Today, there are many sources of data, such as IoT devices, that produce a large amount of data, especially in the healthcare sector. In order to do data analysis, mining, learning analytics, and decision-making tasks as well as for medical research, this microdata must be published and disseminated. But this published data contains sensitive and private information for individuals, and if this microdata is published in its original format, the privacy of individuals could be disclosed, which puts the individuals at risk, especially if the adversary has in-depth prior information of the target person. For this reason, it's crucial for individuals, data providers, and researchers to achieve privacy while improving utility during publishing and mining data. Most of previous works used 1: 1 dataset which contain only one record and one sensitive attribute for individual, however owning multiple records and multiple sensitive attributes (MSA) for an individual can lead to new privacy leakages or disclosure. So, the fundamental issue is how to protect the privacy of 1:M dataset using anonymization techniques and methods, as well as how to balance utility and privacy, for this data, reducing information loss and misuse during mining analytics, in order to make use of anonymized dataset. In this thesis, we present a generic 1: M data privacy and mining model based on the model of (k, l)-diversity, a novel privacy approach that addresses the risk of disclosure in 1: M data publishing. Based on this model, we create an effective technique to preserve privacy and enhance data utility while applying different mining classification algorithms on anonymized data to make sure data utility and contrast it with other methods. Using a variety of real-world datasets and extensive experimentation, it has been shown that our method operates more effectively than the previous techniques in terms of data privacy and utility for mining analysis results.

PHD Title

PHD Abstract