Giving the Keynote talk at HCIR conference on Thursday, October 3.Blurring of the Boundary between Search and Recommendation
Ed H. Chi, Research Scientist, Google
As search became more social and personalized, search engines also have become more interactive. Interestingly, as a comparison, recommendation systems have always been social and personalized, as well as highly interactive, starting from the early Collaborative Filtering work led by my late advisor, John Riedl. But recommendation systems have also increasingly been more search-like by offering capabilities that enable users to tune and direct recommendation results instantly.
As the two technologies evolve toward each other, there is increasingly a blurring of the boundary between these two approaches to interactive information seeking. On the search side, some of this is driven by the merging of question answering capabilities with search, led by systems like Google with Google Now and Apple with Siri that move search toward intelligent personal assistants. On the recommendation side, it is a merging of techniques from not just keyword search but also faceted search with user-based and item-based collaborative filtering techniques and other more proactive recommendation systems.
This blurring has resulted in both critical re-thinking not just about how to architect the systems by merging and sharing backend components common to both types of systems, but also how to structure the user interaction and experience. How can we make progress on these new research problems? I will illustrate my thoughts on these problems by looking at our experience in building (1) a social search engine using Delicious social tagging data, (2) personalized newspaper and conversational recommender for Twitter, and (3) ways to use HCI experimental techniques, such as eyetracking, to understand social explanations for search and recommendation results.keynote | hcirworkshop
/cc +Research at Google