Title :
Building Classifiers with Independency Constraints
Author :
Calders, Toon ; Kamiran, Faisal ; Pechenizkiy, Mykola
Author_Institution :
Eindhoven Univ. of Technol., Eindhoven, Netherlands
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.
Keywords :
learning (artificial intelligence); pattern classification; biased decision process; classification; classifier learning; classifier prediction; classifier training; data attributes; data contains; independency constraints problem; labeling criteria; training data; Conferences; Constraint optimization; Data mining; Electronic mail; Labeling; Machine learning; Machine learning algorithms; Prediction algorithms; Predictive models; Training data;
Conference_Titel :
Data Mining Workshops, 2009. ICDMW '09. IEEE International Conference on
Conference_Location :
Miami, FL
Print_ISBN :
978-1-4244-5384-9
Electronic_ISBN :
978-0-7695-3902-7
DOI :
10.1109/ICDMW.2009.83
