An asymmetric entropy measure for decision trees

Marcellin, Simon; Zighed, Djamel A.; Ritschard, Gilbert

Proceedings chapter

French

An asymmetric entropy measure for decision trees

ContributorsMarcellin, Simon; Zighed, Djamel A.; Ritschard, Gilbert

Presented atParis, France, 2006

Published in11th Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2006, p. 1292-1299

Publication date2006

Abstract

In this paper we present a new entropy measure to grow decision trees. This measure has the characteristic to be asymmetric, allowing the user to grow trees which better correspond to his expectation in terms of recall and precision on each class. Then we propose decision rules adapted to such trees. Experiments have been realized on real medical data from breast cancer screening units.

Keywords

Decision trees
Entropy
Class imbalance
Asymmetric learning
Decision rules

Affiliation entities

Faculté des sciences de la société / Institut de démographie et de socioéconomie

Citation (ISO format)

MARCELLIN, Simon, ZIGHED, Djamel A., RITSCHARD, Gilbert. An asymmetric entropy measure for decision trees. In: 11th Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2006. Paris, France. [s.l.] : [s.n.], 2006. p. 1292–1299.

Proceedings chapter

Identifiers

PID : unige:4531

761views

754downloads

Creation01/12/2009 15:28:07

First validation01/12/2009 15:28:07

Update time14/03/2023 15:19:07

Status update14/03/2023 15:19:07

Last indexation29/10/2024 12:44:56

Archive ouverte UNIGE

An asymmetric entropy measure for decision trees

Technical informations