Publication
Title
Building classifiers with independency constraints
Author
Abstract
In this paper we study the problem of classifier learning where the input data contains unjustified dependencies between some data attributes and the class label. Such cases arise for example when the training data is collected from different sources with different labeling criteria or when the data is generated by a biased decision process. When a classifier is trained directly on such data, these undesirable dependencies will carry over to the classifier's predictions. In order to tackle this problem, we study the classification with independency constraints problem: find an accurate model for which the predictions are independent from a given binary attribute. We propose two solutions for this problem and present an empirical validation.
Language
English
Source (book)
9th IEEE International Conference on Data Mining, December 06-09, 2009, Miami Beach, Florida
Publication
New York, N.Y. : IEEE, 2009
ISBN
978-1-4244-5384-9
Volume/pages
(2009), p. 13-18
ISI
000290247100003
Full text (Publisher's DOI)
UAntwerpen
Faculty/Department
Publication type
Subject
External links
Web of Science
Record
Identification
Creation 23.06.2016
Last edited 03.07.2017