Reference Information:
Title: Agent-Assisted Task Management that Reduces Email Overload
Authors: Andrew Faulring, Brad Myers, Ken Mohnkern, Bradley Schmerl, Aaron Steinfeld, John Zimmerman, Asim Smailagic, Jeffery Hansen, and Daniel Siewiore.
Venue: IUI’10, February 7–10, 2010, Hong Kong, China.
Summary: The authors of this paper have designed and developed a system known as RADAR, or Reflective Agents with Distributed Adaptive Reasoning. RADAR is a email management system that uses a task-based metaphore to attempt to organize a user's inbox. Using machine learning, the system analyzes incoming email to detect if the email is talking about a particular task. Once it has detected a task, it groups all future emails together with the initial email, allowing users to gather emails in a much more logical fashion.
The researchers ran several tests cases with users, and found that the test users on the RADAR system had an easier time finding emails. The researchers were pleasantly surprised at how few false-positives there were with the system.
Discussion: This research was incredibly interesting. It reminded me of Gmail's tagging feature, but not reliant on users to create tags. The overall contents of the article seemed incredibly technical. Overall, it was a good paper none the less.
I feel like I would prefer to use threaded email.
ReplyDeleteI think the system is interesting, but I am not sure if I'd like using it or not.
ReplyDelete