Adding monotonicity to learning algorithms may impair their accuracy

Authors:
Arie Ben-David;Leon Sterling;TriDat Tran
Affiliations:
Management Information Systems, Department of Technology Management, Holon Institute of Technology, 52 Golomb Street, P.O. Box 305, Holon 58102, Israel;Department of Computer Science and Software Engineering, University of Melbourne, Melbourne, Australia;Department of Computer Science and Software Engineering, University of Melbourne, Melbourne, Australia
Venue:
Expert Systems with Applications: An International Journal
Year:
2009

Citing 9
Cited 8

Learning and classification of monotonic ordinal concepts

Computational Intelligence
C4.5: programs for machine learning

C4.5: programs for machine learning
Monotonicity Maintenance in Information-Theoretic Machine Learning Algorithms

Machine Learning
Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
Classification trees for problems with monotonicity constraints

ACM SIGKDD Explorations Newsletter
Optimizing search engines using clickthrough data

Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
New approaches to support vector ordinal regression

ICML '05 Proceedings of the 22nd international conference on Machine learning
Statistical Comparisons of Classifiers over Multiple Data Sets

The Journal of Machine Learning Research
Prediction of Ordinal Classes Using Regression Trees

Fundamenta Informaticae - Intelligent Systems

Two algorithms for generating structured and unstructured monotone ordinal data sets

Engineering Applications of Artificial Intelligence
Sequential covering rule induction algorithm for variable consistency rough set approaches

Information Sciences: an International Journal
Citation-based journal ranks: The use of fuzzy measures

Fuzzy Sets and Systems
Learning monotone nonlinear models using the choquet integral

ECML PKDD'11 Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part III
The discovery and use of ordinal information on attribute values in classifier learning

AI'11 Proceedings of the 24th international conference on Advances in Artificial Intelligence
Performance of classification models from a user perspective

Decision Support Systems
Hierarchical model for rank discrimination measures

ECSQARU'13 Proceedings of the 12th European conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty
Comprehensible classification models: a position paper

ACM SIGKDD Explorations Newsletter

Quantified Score

Hi-index	12.05

Visualization

Abstract

Ordinal (i.e., ordered) classifiers are used to make judgments that we make on a regular basis, both at work and at home. Perhaps surprisingly, there have been no comprehensive studies in the scientific literature comparing the various ordinal classifiers. This paper compares the accuracy of five ordinal and three non-ordinal classifiers on a benchmark of fifteen real-world datasets. The results show that the ordinal classifiers that were tested had no meaningful statistical advantage over the corresponding non-ordinal classifiers. Furthermore, the ordinal classifiers that guaranteed monotonic classifications showed no meaningful statistical advantage over a majority-based classifier. We suggest that the tested ordinal classifiers did not properly utilize the order information in the presence of non-monotonic noise.