Publication
Privacy User Interfaces for Intelligent Retail Stores
Frederic Raber
Privacy berlin 2017 - ITG Workshop on Privacy in Ubiquitous Computing, 2017.
Abstract
Intelligent Retail Stores like Amazon Go or the Innovative Retail Lab in Germany are becoming more and more popular in the last years and will be part of an everyday shopping activity in the next decades, according to current studies. This kind of stores offer services like personalized product recommendations or an “invisible checkout”, where the customer can just leave the store with his shopping bag, without having to pass a checkout system. The bought products are detected by the store environment and withdrawn from the credit card of the registered customer. Although both sides, the customer as well as the retailer can profit from such an ubiquitous shopping system, intelligent retail stores have to capture a lot of sensitive private information like shopper movements or picked-up products to make these convenience services work.
In a first step, we conducted an expert interview and a pilot study to determine which types of private data are recorded in such a retail store, and grouped them with ascending sensitivity into clusters.
We then present a privacy UI, called URetail, that gives back to the customer control over his own data, by offering an easy-to-use interface to select which of his private data should be disclosed. We implemented two interfaces, a traditional list-based interface and a modern radar-based UI. While the former is well-known e.g. from social networks like Facebook, the latter uses a radar metaphor to arrange the permissions with ascending sensitivity into different clusters, and lets the user set the
disclosement policies for each of the clusters with a single click. The final evaluation study shows that both traditional and radar interfaces have their own strengths, although the radar interface has an outstanding usability and user experience compared to the list-based UI.