The Fril fuzzy data browser is a software tool which can automatically derive rules from large bodies of data. The data need not be completely known, and the derived rules can be used to fill in missing values, highlight anomalous values, or predict values in new cases. Human expertise can be input at any stage, and hierarchical systems of rules can be generated. Rules use the fuzzy or evidential logic uncertainty calculus built-in to Fril. It is also possible to generate C-code, although rules are easier to understand, and more efficiently executed in Fril. An enhanced version of the fuzzy data browser is linked to Mathematica, giving access to sophisticated graphical and mathematical facilities. We focus on some simple examples to illustrate the use of the enhanced fuzzy data browser in developing rules which model data.<<ETX>>
[1]
C. Welty,et al.
A Critical Review of Data on Field-Scale Dispersion in Aquifers
,
1992
.
[2]
James F. Baldwin.
Fuzzy rule automation from data using mass assignment theory
,
1995,
Proceedings of 1995 IEEE International Conference on Fuzzy Systems..
[3]
James F. Baldwin,et al.
Fuzzy Sets, Fuzzy Clustering and Fuzzy Rules in AI
,
1993,
Fuzzy Logic in Artificial Intelligence.
[4]
Trevor P Martin,et al.
Refining knowledge from uncertain relations-a fuzzy data browser based on fuzzy object-oriented prog
,
1995
.
[5]
Ian Graham.
FRIL: A Support Logic Programming System
,
1989
.
[6]
Lawrence S. Kroll.
Mathematica--A System for Doing Mathematics by Computer.
,
1989
.