Learning Routines to Support Busy Families
|Scott Davidoff||Carnegie Mellon University||[Home Page]|
Notice: Hosted by Aaron Spaulding
Date: 2011-03-29 at 16:00
Location: EJ228 (SRI E building) (Directions)
As sensing technologies trend towards ubiquity, human activities are becoming accessible to computational pattern recognition. The repetitive nature of routines, and their importance in daily life, makes them a natural subject for these growing computational capabilities. Current research on routine often focuses on event or pattern detection. This highly constrained definition of routine limits our vision of what makes appropriate subjects for modeling, and the desirability of the end-user applications they can support.
Because routines, play a central role to people in their daily lives, however, we can exploit routine as a resource for systems design. A deep understanding of how routines support daily life will uncover new subjects for modeling, which can then introduce new ways to support end-users.
I illustrate how we can leverage routines to create new support applications by focusing on dual-income families. Dual-income families rely on routines to support the detail required to make and monitor transportation plans for kids activities. Successful routines reduce anxiety levels, and provide parents with the feeling of confidence, competence, and control.
The study of how routines support dual-income family coordination creates the possibility for new kinds of support tools. My study of family logistics shows that family members sometimes need but do not have access to information about the plans and routines of other family members. Because family members do not document these routines, they do not exist as a resource family members can turn to when needed.
Using mobile phone GPS, I demonstrate how machine learning and data mining can automatically document those undocumented family transportation routines, generating new resources that family members can turn to when needed. I demonstrate that family members find this new resource useful to their coordination, and that it helps them feel like more competent parents, more in control of their lives.
Scott Davidoff has a PhD in Human-Computer Interaction from Carnegie Mellon, where he was advised by Anind Dey and John Zimmerman. Scott also worked at Microsoft Research Cambridge (UK) with Shahram Izadi and Alex Taylor. Previously, he spent 7 years managing interaction design boutique Scott Davidoff Design, where he developed new products for companies like AOL, SBC Ameritech, and TV Guide. Scott also has an MS in Computer Science (Research) and an M.HCI in Human-Computer Interaction (Practice), both from Carnegie Mellon.
For more information: http://scottdavidoff.com
Please arrive at least 10 minutes early in order to sign in and be escorted to the conference room. SRI is located at 333 Ravenswood Avenue in Menlo Park. Visitors may park in the visitors lot in front of Building E, and should follow the instructions by the lobby phone to be escorted to the meeting room. Detailed directions to SRI, as well as maps, are available from the Visiting AIC web page.