Combining personalization and privacy for user data

Computer scientists and legal experts from Trinity College Dublin and SFI's ADAPT centre are working to marry two of cyberspace's greatest desires, by simultaneously providing enhanced options for user personalisation alongside tighter, more transparent privacy controls around our precious data.

Specifically, they are developing a black-and-white online privacy system, similar to the Creative Commons Licensing system, which could be added to websites and applications so that users signing up would know exactly how private, or otherwise, their personal information would be.

Associate Professor in Law at Trinity, Eoin O'Dell, is spearheading the idea. He said: "On the one hand, we are increasingly told that privacy is dead, that big data is the new oil, and that the privacy costs of highly personalised content are dwarfed by the social and personal benefits of individualised products, services, and content aligned to personal interests and preferences."

"But on the other hand, you don't know what you've lost till it's gone, and this is nowhere more true than where freedom of choice (about participation and privacy) is simply assumed away. The EU is currently working towards a Data Protection Regulation to supply a baseline protection for privacy interests, but lots more must be done to ensure that privacy is respected and protected as we reap the benefits of the information revolution."

Professor O'Dell will be speaking about this project at the 2015 User Modelling, Adaptation and Personalization Conference (UMAP), which takes place in Trinity between Monday, June 29 and Friday, July 3.

Currently termed 'Privacy Paradigm', the goals of the system closely complement the work of Trinity and ADAPT's computer scientists, who are seeking to build personalised visualisations that show users not only what a system has modelled about them, but give the same users simple-to-use controls that can alter and adjust those models. This form of explorative personalisation allows users to reflect on what they are doing, and to identify missed opportunities.

Assistant Professor in Computer Science at Trinity, Owen Conlan, said: "It's a grand target we're setting ourselves and the research is ongoing but the big-picture vision is to make the way online services use our personal – and often privacy sensitive – information as transparent and easy to understand and manipulate as possible for ordinary users."

Innovative technologies such as personalisation, in which Trinity researchers are world leaders, present the opportunity for each piece of content and every interaction offered to a user to be tailored to their individual needs. To do this highly advanced algorithms are employed to model those user needs.

This presents a significant issue – what happens when the algorithm gets it wrong? The personalisation systems are highly advanced, so it can be difficult for users to diagnose what is going on. For example, why does Facebook present posts in the way it does so that you see some and not others from your networks?

"In essence another algorithm is deciding what you see. We in Trinity are putting users back in the driving seat, while ensuring they benefit from the huge potential that such detailed personalisation can offer," added Professor Conlan.