Go to page content

Privacy-preserving Query Processing on Health Data

Friday, May 25, 1-2 p.m.

EN-2022

Seminar given by M.Sc. student Mohammad Sheykholeslam. Due to the huge volume of digital data and the underlying complexity of data management, people and companies are motivated to outsource their computational requirements to the cloud. A significant portion of these productions are used in health applications. While popular cloud computing platforms provide flexible and low-priced solutions, unfortunately, they do so with little support for data security and privacy. This shortcoming clearly threatens sensitive data in cloud platforms. This is especially true for health information, which should always be adequately secured via encryption. Providing secure storage and access to health information that is generated by systems or used in applications, is the main challenge in today’s health care systems. As a result, owners of sensitive information may hesitate in purchasing such services, given the risks associated with the unauthorized access to their data. Considering this problem, researchers have recommended applying encryption algorithms. Data owners never disclose encryption keys in order to keep their encrypted data secure. Because cloud platforms can not search in data which encrypted with regular encryption algorithms, it is supposed that data owners conceal their secrets with searchable encryption algorithms. Searchable encryption is a family of cryptographic protocols that facilitate private keyword searches directly on encrypted data. These protocols allow data owners to upload their encrypted data to the cloud, while retaining the ability to query over uploaded data. In this project, we focus on symmetric searchable encryption schemes, as well as apply an efficient searchable encryption scheme which supports multi-keyword searches to provide a privacy pre serving keyword search framework for health data. Our framework applies a recent secure searchable encryption scheme and employs an inverted indexing structure in order to process queries in a privacy-preserving manner.

Presented by Department of Computer Science

Event Listing 2018-05-25 13:00:00 2018-05-25 14:00:00 America/St_Johns Privacy-preserving Query Processing on Health Data Seminar given by M.Sc. student Mohammad Sheykholeslam. Due to the huge volume of digital data and the underlying complexity of data management, people and companies are motivated to outsource their computational requirements to the cloud. A significant portion of these productions are used in health applications. While popular cloud computing platforms provide flexible and low-priced solutions, unfortunately, they do so with little support for data security and privacy. This shortcoming clearly threatens sensitive data in cloud platforms. This is especially true for health information, which should always be adequately secured via encryption. Providing secure storage and access to health information that is generated by systems or used in applications, is the main challenge in today’s health care systems. As a result, owners of sensitive information may hesitate in purchasing such services, given the risks associated with the unauthorized access to their data. Considering this problem, researchers have recommended applying encryption algorithms. Data owners never disclose encryption keys in order to keep their encrypted data secure. Because cloud platforms can not search in data which encrypted with regular encryption algorithms, it is supposed that data owners conceal their secrets with searchable encryption algorithms. Searchable encryption is a family of cryptographic protocols that facilitate private keyword searches directly on encrypted data. These protocols allow data owners to upload their encrypted data to the cloud, while retaining the ability to query over uploaded data. In this project, we focus on symmetric searchable encryption schemes, as well as apply an efficient searchable encryption scheme which supports multi-keyword searches to provide a privacy pre serving keyword search framework for health data. Our framework applies a recent secure searchable encryption scheme and employs an inverted indexing structure in order to process queries in a privacy-preserving manner. EN-2022 Department of Computer Science