Undoing Statistical Advice

Authors:
Russell Almond
Affiliations:
-
Venue:
IDA '99 Proceedings of the Third International Symposium on Advances in Intelligent Data Analysis
Year:
1999

Citing 5
Cited 0

Garnet: Comprehensive Support for Graphical, Highly Interactive User Interfaces

Computer
Interactive graphic design using automatic presentation knowledge

CHI '94 Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Reusable hierarchical command objects

Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Statistical Models in S

Statistical Models in S
Graphical Models in Applied Multivariate Statistics

Graphical Models in Applied Multivariate Statistics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Statistical advisors are small pieces of code that are run in response to user actions which provide statistical support for the user, especially users who lack statistical sophistication. These could provide advice in the form of a message, or could automatically perform supplemental actions which change the model according to the advice. However, when the original action is undone, the supplemental actions need to be undone as well. This paper describes an implementation of an intelligent advisory system, NAEPVUE, for analyzing data from the National Assessment of Educational Progress (NAEP). NAEPVUE uses an object-oriented data dictionary for storing statistical advisors. Built in the Amulet user interface development environment, NAEPVUE is able to take advantage of the Myers and Kosbie hierarchical undo model to undo the actions of statistical advisors.