Title :
Training Classifiers for Tree-Structured Sets of Categories
Author :
Gutiérrez-González, D. ; Ortega-Moral, M. ; De-Pablo, M.L. ; Cid-Sueiro, J.
Author_Institution :
Dept. of Signal Theor. & Commun., Univ. Carlos III de Madrid
Abstract :
In this paper we propose a new method for training classifiers for multi-class problems when classes are not (necessarily) mutually exclusive and may be related by means of a probabilistic tree structure. Our method is based on the definition of a Bayesian model relating network parameters, feature vectors and categories. Learning is stated as a maximum likelihood estimation problem of the classifier parameters. The proposed algorithm is tested on an image retrieval scenario
Keywords :
Bayes methods; category theory; directed graphs; feature extraction; learning (artificial intelligence); maximum likelihood estimation; pattern classification; tree data structures; trees (mathematics); Bayesian model; classifier training; feature vectors; image retrieval; learning; maximum likelihood estimation; multiclass problems; probabilistic tree structure; Bayesian methods; Classification tree analysis; Graphical models; Image retrieval; Machine learning; Maximum likelihood estimation; Support vector machine classification; Support vector machines; Testing; Tree data structures;
Conference_Titel :
Machine Learning for Signal Processing, 2005 IEEE Workshop on
Conference_Location :
Mystic, CT
Print_ISBN :
0-7803-9517-4
DOI :
10.1109/MLSP.2005.1532916
