PlateClick: Bootstrapping Food Preferences Through an Adaptive Visual Interface.

Food preference learning is an important component of wellness applications and restaurant recommender systems as it provides personalized information for effective food targeting and suggestions. However, existing systems require some form of food journaling to create a historical record of an individual's meal selections. In addition, current interfaces for food or restaurant preference elicitation rely extensively on text-based descriptions and rating methods, which can impose high cognitive load, thereby hampering wide adoption. In this paper, we propose PlateClick, a novel system that bootstraps food preference using a simple, visual quiz-based user interface. We leverage a pairwise comparison approach with only visual content. Using over 10,028 recipes collected from Yummly, we design a deep convolutional neural network (CNN) to learn the similarity distance metric between food images. Our model is shown to outperform state-of-the-art CNN by 4 times in terms of mean Average Precision. We explore a novel online learning framework that is suitable for learning users' preferences across a large scale dataset based on a small number of interactions (≤ 15). Our online learning approach balances exploitation-exploration and takes advantage of food similarities using preference-propagation in locally connected graphs. We evaluated our system in a field study of 227 anonymous users. The results demonstrate that our method outperforms other baselines by a significant margin, and the learning process can be completed in less than one minute. In summary, PlateClick provides a light-weight, immersive user experience for efficient food preference elicitation.