Abstract

AbstractA new method for the induction of fuzzy decision trees is introduced. The fuzzy decision tree classifier improves prediction accuracy using smaller models by locating more robust splitting regions. The proposed method also provides a measure of confidence for sample classification by propagating partition memberships into all leaf nodes, thereby relaxing local subspace restrictions. The fuzzy decision tree algorithm is presented and compared against standard and bagged decision tree classifiers. Copyright © 2003 John Wiley & Sons, Ltd.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call