kernel-target-alignment

A wider documentation is available at Kta/Doc/KTA_doc.pdf

This program aims to improve the performance of a M-SVM by using kernel target alignment. It is applied here to a classification problem : predicting stereochemical variations along a protein sequence. Further information will be added when the article is published.

Bibliography :

Hybrid hierarchical method for predicting protein ω-twist angles (Problem under study)

Yann Guermeur, Thérèse E Malliavin. Méthode hiérarchique hybride de prédiction des angles de torsion ω des protéines. Rencontres de la Société Francophone de Classification (SFC), Pascal Préa, Sep 2024, Marseille (CIRM, Centre International de Rencontres Mathématiques), France. ffhal-04729602f

Target Kernel Alignment

Cristianini, N., Shawe-Taylor, J., Elisseeff, A. & Kandola, J. S. (2002). On Kernel-Target Alignment. In T. G. Dietterich, S. Becker & Z. Ghahramani (eds.), Advances in Neural Information Processing Systems 14 --- Proceedings of the 2001 Neural Information Processing Systems Conference (NIPS 2001), December 3-8, 2001, Vancouver, British Columbia, Canada (p./pp. 367--373), : MIT Press, Cambridge, MA, USA.

M-SVM used

Yann Guermeur and Emmanuel Monfrini. 2011. A Quadratic Loss Multi-Class SVM for which a Radius-Margin Bound Applies. Informatica 22, 1 (January 2011), 73–96.

Weston, Jason & Watkins, Christopher. (1999). Support Vector Machines for Multi-Class Pattern Recognition. Proc of the 7th European Sympo-sium On Artificial Neural Networks. 219-224.

(the code for both M-SVM is available on https://members.loria.fr/YGuermeur/)