Abstract
In this work we investigate several issues in order to improve the per-
formance of probabilistic estimation trees (PETs). First, we derive a new prob-
ability smoothing that takes into account the class distributions of all the nodes
from the root to each leaf. Secondly, we introduce or adapt some new splitting
criteria aimed at improving probability estimates rather than improving classifi-
cation accuracy, and compare them with other accuracy-aimed splitting criteria.
Thirdly, we analyse the effect of pruning methods and we choose a cardinality-
based pruning, which is able to significantly reduce the size of the trees without
degrading the quality of the estimates. The quality of probability estimates of
these three issues is evaluated by the 1-vs-1 multi-class extension of the Area
Under the ROC Curve (AUC) measure, which is becoming widespread for
evaluating probability estimators, ranking of predictions in particular.
Translated title of the contribution | Improving the AUC of Probabilistic Estimation Trees |
---|---|
Original language | English |
Title of host publication | Proceedings of the 14th European Conference on Machine Learning |
Pages | 121-132 |
Publication status | Published - 2003 |
Bibliographical note
ISBN: 3540201211Publisher: Springer-Verlag
Name and Venue of Conference: Proceedings of the 14th European Conference on Machine Learning
Other identifier: 2000549