Balancing efficiency and interpretability in an interactive statistical assistant

ABSTRACT

Making an interface more efficient, in a task analysis sense, can make it more difficult for an automated reasoning system to infer user goals, by eliminating some user actions, by presenting information without requiring overt user selection, and so forth. We call the extent to which a system can make such inferences interpretability. In this paper we describe the tradeoff between interpretability and efficiency. We give some general heuristics for improving interpretability in a system and explain how they apply in an implemented system, an assistant for exploratory statistical analysis. Increased interpretability in the system is provided by navigation techniques for data exploration and a data mountain for organizing results; a formative evaluation illustrates some of the potential benefits of applying interpretability heuristics to an intelligent user interface

REFERENCES

• 1: Allen Cypher, Eager: programming repetitive tasks by demonstration, Watch what I do: programming by demonstration, MIT Press, Cambridge, MA, 1993
• 2: D. Christopher Dryer, Wizards, guides, and beyond: rational and empirical methods for selecting optimal intelligent user interface agents, Proceedings of the 2nd international conference on Intelligent user interfaces, p.265-268, January 06-09, 1997, Orlando, Florida, United States [doi>10.1145/238218.238347]
• 3: Richard W. Hamming, Numerical Methods for Scientists and Engineers, McGraw-Hill, Inc., New York, NY, 1973
• 4: H. Lieberman. Letizia: An agent that assists web browsing. In Proceedings of the International Joint Conference on Artificial Intelligence, pages 924--929, 1995.
• 5: David Maulsby , Saul Greenberg , Richard Mander, Prototyping an intelligent agent through Wizard of Oz, Proceedings of the SIGCHI conference on Human factors in computing systems, p.277-284, April 24-29, 1993, Amsterdam, The Netherlands [doi>10.1145/169059.169215]
• 6: Richard McDaniel, Demonstrating the hidden features that make an application work, Your wish is my command: programming by example, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2001
• 7: Brad A. Myers , David S. Kosbie, Reusable hierarchical command objects, Proceedings of the SIGCHI conference on Human factors in computing systems: common ground, p.260-267, April 13-18, 1996, Vancouver, British Columbia, Canada [doi>10.1145/238386.238526]
• 8: Mark O. Riedl , Robert St. Amant, Toward automated exploration of interactive systems, Proceedings of the 7th international conference on Intelligent user interfaces, January 13-16, 2002, San Francisco, California, USA [doi>10.1145/502716.502738]
• 9: George Robertson , Mary Czerwinski , Kevin Larson , Daniel C. Robbins , David Thiel , Maarten van Dantzich, Data mountain: using spatial memory for document management, Proceedings of the 11th annual ACM symposium on User interface software and technology, p.153-162, November 01-04, 1998, San Francisco, California, United States [doi>10.1145/288392.288596]
• 10: Robert St. Amant , Paul R. Cohen, Interaction with a mixed-initiative system for exploratory data analysis, Proceedings of the 2nd international conference on Intelligent user interfaces, p.15-22, January 06-09, 1997, Orlando, Florida, United States [doi>10.1145/238218.238221]
• 11: R. St. Amant and P. R. Cohen. Intelligent support for exploratory data analysis. Journal of Computational and Graphical Statistics, 7(4):545--558, 1998.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.2 User Interfaces (D.2.2, H.1.2, I.3.6)

General Terms:
Human Factors

Keywords:
data mountain, efficiency, interpretability, navigation, user interface design