A Caltech Library Service

The Multilevel Classification Problem and a Monotonicity Hint

Magdon-Ismail, Malik and Chen, Hung-Ching and Abu-Mostafa, Yaser S. (2002) The Multilevel Classification Problem and a Monotonicity Hint. In: Intelligent Data Engineering and Automated Learning - IDEAL 2002. Lecture Notes in Computer Science. No.2412. Springer , Berlin, pp. 410-415. ISBN 9783540440253.

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item:


We introduce and formalize the multilevel classification problem, in which each category can be subdivided into different levels. We analyze the framework in a Bayesian setting using Normal class conditional densities. Within this framework, a natural monotonicity hint converts the problem into a nonlinear programming task, with non-linear constraints. We present Monte Carlo and gradient based techniques for addressing this task, and show the results of simulations. Incorporation of monotonicity yields a systematic improvement in performance.

Item Type:Book Section
Related URLs:
URLURL TypeDescription ReadCube access
Additional Information:© 2002 Springer-Verlag Berlin Heidelberg. First Online 20 August 2002. Many have contributed to the progress of this work. In particular, we single out Honeywell Corporation for alerting us to the problem and providing initial motivation, James Psota and Amir Atiya for useful discussion.
Subject Keywords:Pattern Recognition Problem; Monotonicity Constraint; Risk Matrix; Bayesian Setting; Gradient Base Approach
Series Name:Lecture Notes in Computer Science
Issue or Number:2412
Record Number:CaltechAUTHORS:20190702-152246968
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:96896
Deposited By: Tony Diaz
Deposited On:08 Jul 2019 16:51
Last Modified:16 Nov 2021 17:24

Repository Staff Only: item control page