An Improved SVM-KM Model for Imbalanced Datasets

Support vector machine is a widely used machine learning technique. SVM-KM model can speed SVM training by eliminating non support vectors, but imbalanced datasets will affect the classification accuracy. In this paper, we proposed an improved SVM-KM model, which assign different error costs to different classes. Based on the simulation results, the improved SVM-KM model performed best for imbalanced datasets.

[1]  Haibo He,et al.  Learning from Imbalanced Data , 2009, IEEE Transactions on Knowledge and Data Engineering.

[2]  Weiguo Gong,et al.  Multi-objective uniform design as a SVM model selection tool for face recognition , 2011, Expert Syst. Appl..

[3]  Song Ji,et al.  An effective algorithm for inverse problem of SVM based on MM algorithm , 2009, 2009 International Conference on Machine Learning and Cybernetics.

[4]  Fang-Xiang Wu,et al.  SVM-RFE based feature selection for tandem mass spectrum quality assessment , 2011, Int. J. Data Min. Bioinform..

[5]  Vasile Palade,et al.  FSVM-CIL: Fuzzy Support Vector Machines for Class Imbalance Learning , 2010, IEEE Transactions on Fuzzy Systems.

[6]  Andreas Wagner,et al.  Instantaneous‐Shape Sampling for Calculating the Electromagnetic Dipole Strength in Transitional Nuclei , 2009 .

[7]  Stephen Kwek,et al.  Applying Support Vector Machines to Imbalanced Datasets , 2004, ECML.

[8]  Begüm Demir,et al.  Clustering-Based Extraction of Border Training Patterns for Accurate SVM Classification of Hyperspectral Images , 2009, IEEE Geoscience and Remote Sensing Letters.

[9]  Yasutoshi Yajima,et al.  Ranking and selecting terms for text categorization via SVM discriminate boundary , 2005 .

[10]  Nello Cristianini,et al.  Controlling the Sensitivity of Support Vector Machines , 1999 .

[11]  Chun-Chin Hsu,et al.  MDS: a novel method for class imbalance learning , 2009, ICUIMC '09.

[12]  Jung-Hsien Chiang,et al.  Hierarchically SVM classification based on support vector clustering method and its application to document categorization , 2007, Expert Syst. Appl..

[13]  Antônio de Pádua Braga,et al.  SVM-KM: speeding SVMs learning with a priori cluster selection and k-means , 2000, Proceedings. Vol.1. Sixth Brazilian Symposium on Neural Networks.

[14]  Y.H. Chen,et al.  Cluster-based support vector machines in text-independent speaker identification , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[15]  Li Xiao,et al.  A Chinese Web Page Classifier Based on Support Vector Machine and Unsupervised Clustering , 2001 .