Agent-assisted task management that reduces email overload

RADAR is a multiagent system with a mixed-initiative user interface designed to help office workers cope with email overload. RADAR agents observe experts to learn models of their strategies and then use the models to assist other people who are working on similar tasks. The agents' assistance helps a person to transition from the normal email-centric workflow to a more efficient task-centric workflow. The Email Classifier learns to identify tasks contained within emails and then inspects new emails for similar tasks. A novel task-management user interface displays the found tasks in a to-do list, which has integrated support for performing the tasks. The Multitask Coordination Assistant learns a model of the order in which experts perform tasks and then suggests a schedule to other people who are working on similar tasks. A novel Progress Bar displays the suggested schedule of incomplete tasks as well as the completed tasks. A large evaluation demonstrated that novice users confronted with an email overload test performed significantly better (a 37% better overall score with a factor of four fewer errors) when assisted by the RADAR agents.

[1]  Michael Gamon,et al.  Task-Focused Summarization of Email , 2004 .

[2]  Brad A. Myers,et al.  Citrine: providing intelligent copy-and-paste , 2004, UIST '04.

[3]  Scott E. Fahlman,et al.  Marker-Passing Inference in the Scone Knowledge-Base System , 2006, KSEM.

[4]  Yiming Yang,et al.  Robustness of adaptive filtering methods in a cross-benchmark evaluation , 2005, SIGIR '05.

[5]  Brad A. Myers,et al.  The Design and Evaluation of User Interfaces for the RADAR Learning Personal Assistant , 2009, AI Mag..

[6]  Ian Smith,et al.  Quality Versus Quantity: E-Mail-Centric Task Management and Its Relation With Overload , 2005, Hum. Comput. Interact..

[7]  Ben Shneiderman,et al.  Direct manipulation vs. interface agents , 1997, INTR.

[8]  Victoria Bellotti,et al.  Managing Activities with TV-ACTA: TaskVista and Activity- Centered Task Assistant , 2006 .

[9]  John Zimmerman,et al.  Vio: a mixed-initiative approach to learning and automating procedural update tasks , 2007, CHI.

[10]  Bradley R. Schmerl,et al.  The Radar Architecture for Personal Cognitive Assistance , 2007, Int. J. Softw. Eng. Knowl. Eng..

[11]  Eric Horvitz,et al.  Principles of mixed-initiative user interfaces , 1999, CHI '99.

[12]  Yiming Yang,et al.  A scalability analysis of classifiers in text categorization , 2003, SIGIR.

[13]  Eric Horvitz,et al.  Attention-Sensitive Alerting , 1999, UAI.

[14]  James Carpenter,et al.  Towel: Towards an Intelligent To-Do List , 2007, Interaction Challenges for Intelligent Assistants.

[15]  Eric Nyberg,et al.  Integrating a Natural Language Message Pre-Processor with UIMA , 2008 .

[16]  Victoria Bellotti,et al.  E-mail as habitat: an exploration of embedded personal information management , 2001, INTR.

[17]  Stephen F. Smith,et al.  Linear Relaxation Techniques for Task Management in Uncertain Settings , 2008, ICAPS.

[18]  Candace L. Sidner,et al.  Email overload: exploring personal information management of email , 1996, CHI.

[19]  Michael Freed,et al.  RADAR: A Personal Assistant that Learns to Reduce Email Overload , 2008, AAAI.

[20]  Alexander I. Rudnicky,et al.  Summarizing non-textual events with 'Briefing' focus , 2007, RIAO.

[21]  Eugene Fink,et al.  Scheduling with Uncertain Resources: Collaboration with the User , 2006, 2006 IEEE International Conference on Systems, Man and Cybernetics.

[22]  Jacek Gwizdka,et al.  TaskView: design and evaluation of a task-based email interface , 2002, CASCON.

[23]  Jacek Gwizdka,et al.  Email in personal information management , 2006, CACM.

[24]  Daniel G. Bobrow,et al.  What a to-do: studies of task management towards the design of a personal task list manager , 2004, CHI.

[25]  Deborah L. McGuinness,et al.  An Intelligent Personal Assistant for Task and Time Management , 2007, AI Mag..

[26]  John Zimmerman,et al.  Survey measures for evaluation of cognitive assistants , 2007 .

[27]  Aaron Steinfeld,et al.  Evaluation of an integrated multi-task machine learning system with humans in the loop , 2007 .