Multiclass SVM active learning algorithm based on decision directed acyclic graph and one versus one

The classical training algorithms of support vector machines (SVM) are supervised learning algorithms which based on large-scale labeled samples, while these labeled samples are not easy to be acquired or labeled costly and class–unbalanced dataset, meanwhile these SVM algorithms are originally designed for the solution of two-class problems. To solve these problems of SVM, An Active learning algorithm based on decision directed acyclic graph (DDAG) for SVM is proposed in the paper, which train the multiclass SVMs using as few labeled instances as possible while maintaining the same SVM performance, or achieving the generalization performance of SVM classification as good as possible. The experimental results on the UCI data show that the proposed approach can achieve higher clasification accuracy, but using less labeled samples, while improving generalization performance and ruducing the marking costs of SVM training.

[1]  Daphne Koller,et al.  Active learning: theory and applications , 2001 .

[2]  Nikolaos Papanikolopoulos,et al.  Multi-class active learning for image classification , 2009, CVPR.

[3]  Klaus Brinker On Multiclass Active Learning with Support Vector Machines , 2004, ECAI.

[5]  K. Bennett,et al.  A support vector machine approach to decision trees , 1998, 1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227).

[6]  Ting Wang,et al.  Active Learning for Online Spam Filtering , 2008, AIRS.

[7]  Kamel Faraoun,et al.  Genetic Programming Approach for Multi-Category Pattern Classification Applied to Network Intrusions Detection , 2006, Int. Arab J. Inf. Technol..

[8]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[9]  Hua Xian-sheng Video genre categorization using SVM classifiers with active learning , 2009 .

[10]  Quan Wen Incremental training algorithm of SVM based on active learning , 2010 .

[11]  Derek Greene,et al.  Score Normalization and Aggregation for Active Learning in Multi-label Classification , 2010 .

[12]  Ran El-Yaniv,et al.  Online Choice of Active Learning Algorithms , 2003, J. Mach. Learn. Res..

[13]  Nello Cristianini,et al.  Large Margin DAGs for Multiclass Classification , 1999, NIPS.

[14]  Kristen Grauman,et al.  What's it going to cost you?: Predicting effort vs. informativeness for multi-label image annotations , 2009, CVPR.

[15]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[16]  Hsuan-Tien Lin,et al.  A note on Platt’s probabilistic outputs for support vector machines , 2007, Machine Learning.

[17]  Pietro Perona,et al.  Entropy-based active learning for object recognition , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[18]  Chih-Jen Lin,et al.  A Comparison of Methods for Multi-class Support Vector Machines , 2015 .

[19]  Feng Chong Organization Names Recognition with Active Learning , 2006 .

[20]  Xian-Sheng Hua,et al.  Two-Dimensional Multilabel Active Learning with an Efficient Online Adaptation Model for Image Classification , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Ashish Kapoor,et al.  Active learning for large multi-class problems , 2009, CVPR.

[22]  Thomas G. Dietterich,et al.  Solving Multiclass Learning Problems via Error-Correcting Output Codes , 1994, J. Artif. Intell. Res..

[23]  Rong Yan,et al.  Multi-class active learning for video semantic feature extraction , 2004, 2004 IEEE International Conference on Multimedia and Expo (ICME) (IEEE Cat. No.04TH8763).

[24]  Chih-Jen Lin,et al.  Probability Estimates for Multi-class Classification by Pairwise Coupling , 2003, J. Mach. Learn. Res..

[25]  Zheng Chen,et al.  Effective multi-label active learning for text classification , 2009, KDD.

[26]  Rong Yan,et al.  Automatically labeling video data using multi-class active learning , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.