Some brief highlights from the 13th ACM Conference on Ubiquitous Computing (Beijing, China, September 2011).

Highlights:

• Check out the conference programme here.
• Some cool tutorials were held; including one on urban sensing by Francesco Calabrese (IBM Dublin/MIT).
• There were a number of workshops: a workshop on crowdsourcing, on research in the large, on Mobile Location Based Services, on Context-Awareness for Self-Managing Systems, the 1st International Symposium on Social and Community Intelligence
• All talks were filmed and are currently available online on slideshare; I particularly recommend watching the panel videos (e.g., Greg Abowd talking about the identity crisis of Ubicomp - more on that from me later).

One of the particular features of conferences that I take time to record is what papers/talks/workshops discussed any kind of recommender systems, which are making an appearance at every single conference that I’ve been to recently. Here is that list for Ubicomp 2011. The highlight of this list (for me) is that none of the recommendation work that appeared here deals with the “traditional” web-settings and (movie!) data sets.

• Workshop Paper (Research in the Large Workshop): Face to Face Makes a Difference – Recommendation Practices of Users of Mobile Services
• Keynote (Mobile Location Based Services Workshop): Learning Techniques in Social and Location-Based Service Recommendation
• Workshop Paper (Symposium on Social and Community Intelligence): Real-time Detection and Recommendation of Thermal Spots by Sensing Collective Behaviors in Paragliding, _**Real-time, participatory sensing in order to recommend thermal hot spots to paragliders (which they need in order to gain altitude).
• Paper (Main conference): Where to Find My Next Passenger? - A recommender system for taxi drivers to find passengers/for passengers to find available taxis
• Paper (Main conference): When Recommendation Meets Mobile: Contextual and Personalized Recommendation On The Go “context-aware and personalized entity recommendation which understands the implicit intent without any explicit user input on the phone”

These are just a sample of papers that I chose based on the fact that “recommendation” appears in their title. If #recsys is your thing, don’t hesitate to check out the rest of the programme as well: in fact, a lot of the papers at the conference were about predicting user behaviour (e.g., predicting home or room occupancy in order to build smart-heating systems); which is, of course, a core component of building a system to make recommendations.