Experience with a learning personal assistant

Personal software assistants that help users with tasks like finding information, scheduling calendars, or managing work-flow will require significant customization to each individual user. For example, an assistant that helps schedule a particular user’s calendar will have to know that user’s scheduling preferences. This paper explores the potential of machine learning methods to automatically create and maintain such customized knowledge for personal software assistants. We describe the design of one particular learning assistant: a calendar manager, called CAP (Calendar APprentice), that learns user scheduling preferences from experience. Results are summarized from approximately five user-years of experience, during which CAP has learned an evolving set of several thousand rules that characterize the scheduling preferences of its users. Based on this experience, we suggest that machine learning methods may play an important role in future personal software assistants.

[1]  Nicholas Negroponte,et al.  The Architecture Machine: Toward a More Human Environment , 1973 .

[2]  R. Rosner.  Computer software , 1978, Nature.

[3]  Tom M. Mitchell,et al.  LEAP: A Learning Apprentice for VLSI Design , 1985, IJCAI.

[4]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[5]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[6]  E. R. Bareiss,et al.  Protos: An Exemplar-Based Learning Apprentice1 , 1987 .

[7]  Stuart J. Russell.  Analogical and inductive reasoning , 1987 .

[8]  J. Ross Quinlan,et al.  Generating Production Rules from Decision Trees , 1987, IJCAI.

[9]  Ray Bareiss,et al.  Protos: An Exemplar-Based Learning Apprentice , 1988, Int. J. Man Mach. Stud..

[10]  Gerard Salton,et al.  Improving retrieval performance by relevance feedback , 1997, J. Am. Soc. Inf. Sci..

[11]  Yasushi Nakauchi,et al.  Groupware that learns , 1991, [1991] IEEE Pacific Rim Conference on Communications, Computers and Signal Processing Conference Proceedings.

[12]  Tom M. Mitchell,et al.  A Personal Learning Apprentice , 1992, AAAI.

[13]  Norman Sadeh,et al.  Look-ahead techniques for micro-opportunistic job shop scheduling , 1992 .

[14]  Susan T. Dumais,et al.  Personalized information delivery: an analysis of information filtering methods , 1992, CACM.

[15]  Tom M. Mitchell,et al.  Interfaces That Learn: A Learning Apprentice for Calendar Management , 1993 .

[16]  Siegfried Bocionek,et al.  Dialog-Based Learning (DBL) for Adaptive Interface Agents and Programming-by-Demonstration Systems , 1993 .

[17]  Pattie Maes,et al.  Evolving agents for personalized information filtering , 1993, Proceedings of 9th IEEE Conference on Artificial Intelligence for Applications.

[18]  Rich Caruana,et al.  Greedy Attribute Selection , 1994, ICML.