CISUC

Two-level hierarchical hybrid SVM-RVM classification model

Authors

Abstract

Support Vector Machines (SVM) and Relevance Vector Machines (RVM) constitute two state-of-the-art learning machines that are currently focus of cutting-edge research. SVM present accuracy and complexity preponderance, but are surpassed by RVM when probabilistic outputs or kernel selection come to discussion.

We propose a two-level hierarchical hybrid SVM-RVM model to combine the best of both learning machines. The proposed model first level uses an RVM to determine the less confident classified examples and the second level then makes use of an SVM to learn and classify the tougher examples. We show the benefits of the hierarchical approach on a text classification task, where the two-levels outperform both learning machines.

Subject

Text mining; RVM; SVM

Related Project

CATCH - Inductive Inference for Large Scale Data Bases Text CATegorization

Conference

IEEE ICMLA 2006, December 2006


Cited by

Year 2011 : 2 citations

 Further Results on Alzheimer Disease Detection on Structural MRI Features
M Conde… - Soft Computing Models in Industrial and …, 2011 - Springer

 Li, W.a b , Miao, D.a , Wang, W.a b
Two-level hierarchical combination method for text classification
(2011) Expert Systems with Applications, 38 (3), pp. 2030-2039.

Year 2010 : 2 citations

 Two-level hierarchical combination method for text classification, W Li, D Miao,Expert Systems with Applications, 2010 - Elsevier

 Hybrid SVM/CART classification of pathogenic species of bacterial meningitis with surface-enhanced Raman scattering, CY Huang, TH Tsai, BC Wen, CW Chung… - … (BIBM), 2010 IEEE

Year 2007 : 2 citations

 Boosted Bayesian Kernel Classifier Method for Face Detection
Tashk, Ali Reza Bayesteh; Faez, Karim;
Third International Conference on Natural Computation, 2007. ICNC 2007. Volume 1, 24-27 Aug. 2007 Page(s):533 - 537.

 Face Detection Using Adaboosted RVM-based Component Classifier
ARB Tashk, A Sayadiyan, SM Valiollahzadeh - Image and Signal Processing and Analysis, 2007. ISPA 2007