Weighted support vector machine based on association rules

This paper presents a weighted support vector machine (WSVM) based on association rules for two-class classification problems. The basic idea of the WSVM is to assign different weights to different data points to minimize impacts of outliers. In this paper, we apply association rules to generate weights to prevent bias to the majority class for imbalanced binary classification problems. Experimental results indicate that the proposed method yields a better generalization in comparison to the standard support vector machines.

[1]  Xuegong Zhang,et al.  Using class-center vectors to build support vector machines , 1999, Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468).

[2]  David R. Musicant,et al.  Successive overrelaxation for support vector machines , 1999, IEEE Trans. Neural Networks.

[3]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machines , 2002 .

[4]  N. Cristianini.  Kernel Methods (Kernel‐Based Learning Methods, Kernel Machine) , 2004 .

[5]  Yue Wang,et al.  Weighted support vector machine for data classification , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[6]  Chih-Jen Lin,et al.  Training v-Support Vector Classifiers: Theory and Algorithms , 2001, Neural Computation.

[7]  Xing Zhang,et al.  A new approach to classification based on association rule mining , 2006, Decis. Support Syst..

[8]  Yanqing Zhang,et al.  Granular support vector machines with association rules mining for protein homology prediction , 2005, Artif. Intell. Medicine.

[9]  Glenn Fung,et al.  Proximal support vector machine classifiers , 2001, KDD '01.

[10]  Ming-Wei Chang,et al.  Leave-One-Out Bounds for Support Vector Regression Model Selection , 2005, Neural Computation.

[11]  Wang Wen-jian.  Granular SVM Learning Algorithm for Processing Imbalanced Data , 2010 .