Title: Recommender Systems Seen Through the Lens of Choice Architecture
Abstract: “How do people make choices?” “How can we help them make better choices?” It’s helpful to have compact, coherent answers to these questions if we want to build recommender systems that support choice processes. This talk begins with a brief summary of the ASPECT and ARCADE models (introduced in “Choice Architecture for Human-Computer Interaction”), which answer these questions. It then uses this framework to shed new light on a sample of subtle questions such as: “How can explanations of recommendations help people make better choices?” and “How can recommender systems help people choose via trial and error?” The talk is a concrete and selective presentation of key ideas from the chapter “Human Decision Making and Recommender Systems” in the second edition of the “Recommender Systems Handbook”.
Slides: Are available here.
Bio: Anthony Jameson is a Principal Researcher at DFKI, the German Research Center for Artificial Intelligence. He has conducted research on interactive intelligent systems in a number of areas in the intersection of artificial intelligence and psychology, including recommender systems. With the late John Riedl, he cofounded the ACM Transactions on Interactive Intelligent Systems. He is lead author of the monograph Choice Architecture for Human-Computer Interaction in Foundations & Trends in Human-Computer Interaction.