Privacy and Big Data
This seminar is about new and emerging approaches to adjust and balance privacy and utility in data intensive applications, such as information retrieval, data mining and personalisation. These new approaches have the potential to enable a new generation of privacy-enabled services which are not focused on maximizing the collection of user data. Instead these new approaches enable user privacy under different threat models, such as protecting the identity of individual users when querying aggregated data, or preventing leakage of query patterns when users retrieve data from a database. As a result, these new approaches may help businesses in their compliance with increasingly regulatory trust and reinforce user trust, while enabling new business models at the same time.
Users of contemporary information technology services have come to expect a “smart” user experience, which requires collecting a lot of data about the user. For instance, Google Maps "knows" about traffic jams because it collects movement data by default from all smart phones it is installed on. In addition, the dominant business model on the web is to offer services for free and monetize the service by selling the user data . Therefore the majority of current services have not been designed with the privacy of users in mind.
Planned topics for the seminar include differential privacy and privacy-enabled data mining; private information retrieval and oblivious data structures / oblivious transfer; blockchain assisted computation; and secure multiparty computation.
This seminar will be organised as a block event ("Blockseminar").