Thumbnail
Access Restriction
Subscribed

Author Berthold, M.R. ♦ Holve, R.
Source IEEE Xplore Digital Library
Content type Text
Publisher Institute of Electrical and Electronics Engineers, Inc. (IEEE)
File Format PDF
Copyright Year ©2000
Language English
Subject Domain (in DDC) Computer science, information & general works ♦ Special computer methods
Subject Keyword Fuzzy sets ♦ Data mining ♦ Euclidean distance ♦ Equations ♦ Data visualization ♦ Iris ♦ Springs ♦ Multidimensional systems ♦ Marine vehicles
Abstract In this paper we present an approach to visualize a potentially high-dimensional and large number of (fuzzy) rules in two dimensions. This visualization presents the entire set of rules to the user as one coherent picture. We use a gradient descent based algorithm to generate a 2D-view of the rule set which minimizes the error on the pair-wise fuzzy distances between all rules. This approach is superior to a simple projection and also most non-linear transformations in that it concentrates on the important feature, that is the inter-point distances. In order to make use of the uncertain nature of the underlying fuzzy rules, a new fuzzy distance-measure was developed. The visualizations of a rule set for the well-known IRIS dataset as well as fuzzy models for other benchmark data sets are illustrated and discussed.
Description Author affiliation: Dept. of Electr. Eng. & Comput. Sci., California Univ., Berkeley, CA, USA (Berthold, M.R.)
ISBN 0780362748
Educational Role Student ♦ Teacher
Age Range above 22 year
Educational Use Research ♦ Reading
Education Level UG and PG
Learning Resource Type Article
Publisher Date 2000-07-13
Publisher Place USA
Rights Holder Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Size (in Bytes) 412.12 kB
Page Count 5
Starting Page 64
Ending Page 68


Source: IEEE Xplore Digital Library