基于类分布的领域自适应支持向量机

应文豪; 王士同; 邓赵红; 王骏

doi:10.3724/SP.J.1004.2013.01273

基于类分布的领域自适应支持向量机

doi: 10.3724/SP.J.1004.2013.01273

1.
江南大学数字媒体学院无锡 214122;
2.
常熟理工学院计算机科学与工程学院常熟 215500

基金项目:

国家自然科学基金(60975027, 61170122);江苏省自然科学基金(BK 2011003)资助

详细信息

作者简介:
王士同江南大学数字媒体学院教授．主要研究方向为人工智能,模式识别和生物信息.E-mail: wxwangst@yahoo.com.cn

计量
- 文章访问数: 1795
- HTML全文浏览量: 57
- PDF下载量: 1199
- 被引次数: 0
出版历程
- 收稿日期: 2012-03-16
- 修回日期: 2012-07-15
- 刊出日期: 2013-08-20

Support Vector Machine for Domain Adaptation Based on Class Distribution

1.
School of Digital Media, Jiangnan University, Wuxi 214122;
2.
School of Computer Science and Engineering, Changshu Institute of Technology, Changshu 215500

Funds:

Supported by National Natural Science Foundation of China (60975027, 61170122), and Natural Science Foundation of Jiangsu Province (BK2011003)

摘要

摘要: 现有的领域自适应方法在定义领域间分布距离时, 通常仅从领域样本的整体分布上考虑, 而未对带类标签的领域样本分布分别进行考虑, 从而在一些具有非平衡数据集的应用领域上表现出一定的局限性. 对此, 在充分考虑源领域样本类信息的基础上, 基于结构风险最小化模型, 提出了基于类分布的领域自适应支持向量机(Domain adaptation support vector machine based on class distribution, CDASVM), 并将其拓展为可处理多源问题的多源领域自适应支持向量机(CDASVM from multiple sources, MSCDASVM), 在人造和真实的非平衡数据集上的实验结果表明, 所提方法具有优化或可比较的模式分类性能.
- 领域自适应 /
- 支持向量机 /
- 迁移学习 /
- 再生核Hilbert空间
Abstract: Current domain adaptation methods almost focus on the whole domain sample's distribution and ignore the sample's label information when they consider the distribution discrepancy between source domain and target domain. So, these methods may not work well on imbalanced datasets. In the paper, we employ the proposed distribution discrepancy which considers the sample's label information in the source domain and then propose a novel domain adaptation learning method based on the structure risk minimization principle, called support vector machine for domain adaptation based on class distribution (CDASVM). Accordingly, the CDASVM is extended to MSCDASVM (CDASVM from multiple sources) which can be used to deal with the domain adaptation problem from multiple sources. Experimental results on artificial and real imbalanced datasets show that the proposed machines CDASVM and MSCDASVM outperform or are comparable to the related domain adaption methods.
- Domain adaptation /
- support vector machine (SVM) /
- transfer learning /
- reproduced kernel Hilbert space

HTML全文

参考文献(29)

[1]	Pan S J, Yang Q. A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359
[2]	Quanz B, Huan J. Large margin transductive transfer learning. In: Proceedings of the 18th ACM conference on Information and knowledge management. New York, USA: ACM, 2009. 1327-1336
[3]	Pan S J, Tsang I W, Kwok J T, Yang Q. Domain adaptation via transfer component analysis. IEEE Transactions on Neural Networks, 2011, 22(2): 199-210
[4]	Xiang E W, Cao B, Hu D H, Yang Q. Bridging domains using world wide knowledge for transfer learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(6): 770-783
[5]	Bruzzone L, Marconcini M. Domain adaptation problems: a DASVM classification technique and a circular validation strategy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(5): 770-787
[6]	Ben-David S, Blitzer J, Crammer K, Pereira F. Analysis of representations for domain adaptation. In: Proceedings of the 2007 Advances in Neural Information Processing Systems 19. Cambridge: MIT Press, 2007. 137-144
[7]	Daumé H III, Marcu D. Domain adaptation for statistical classifiers. Journal of Artificial Intelligence Research, 2006, 26(1): 101-126
[8]	Ling X, Dai W Y, Xue G R, Yang Q, Yu Y. Spectral domain-transfer learning. In: Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. New York, NY, USA: ACM, 2008. 488-496
[9]	Dai W Y, Xue G R, Yang Q, Yu Y. Co-clustering based classification for out-of-domain documents. In: Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM, 2007: 210-219
[10]	Blitzer J, Dredze M, Pereira F. Biographies, bollywood, boom-boxes and blenders: domain adaptation for sentiment classification. In: Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics. Prague, Czekh Republic: Association for Computational Linguistics, 2007. 440-447
[11]	Sriperumbudur B K, Gretton A, Fukumizu K, Schölkopf, Lanckriet G R G. Hilbert space embeddings and metrics on probability measures. Journal of Machine Learning Research, 2010, 11: 1517-1561
[12]	Sriperumbudur B K, Fukumizu K, Gretton A, Lanckriet G R G, Schölkopf. Kernel choice and classifiability for RKHS embeddings of probability distributions. In: Proceedings of the 2009 Advances in Neural Information Processing Systems. Cambridge, MA: MIT Press, 2009, 22: 1750-1758
[13]	Gretton A, Fukumizu K, Harchaoui Z, Sriperumbudur B K. A fast, consistent kernel two-sample test. In: Proceedings of the 2010 Advances in Neural Information Processing Systems. Cambridge, MA: MIT Press, 2010. 673-681
[14]	Vapnik V N. The Nature of Statistical Learning Theory. New York: Springer-Verlag, 1995. 123-167
[15]	Pal M, Foody G M. Feature selection for classification of hyper spectral data by SVM. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(5): 2297-2307
[16]	Liu Jian-Wei, Li Shuang-Cheng, Luo Xiong-Lin. Classification algorithm of support vector machine via p-norm regularization. Acta Automatica Sinica, 2012, 38(1): 76-87(刘建伟, 李双成, 罗雄麟. p范数正则化支持向量机分类算法. 自动化学报, 2012, 38(1): 76-87)
[17]	Liu Qiao, Qin Zhi-Guang, Chen Wei, Zhang Feng-Li. Zero-norm penalized feature selection support vector machine. Acta Automatica Sinica, 2011, 37(2): 252-256(刘峤, 秦志光, 陈伟, 张凤荔. 基于零范数特征选择的支持向量机模型. 自动化学报, 2011, 37(2): 252-256)
[18]	Hu Wen-Jun, Wang Shi-Tong. Fast real-time decision approach of support vector data description. Acta Automatica Sinica, 2011, 37(9): 1085-1094(胡文军, 王士同. SVDD的快速实时决策方法. 自动化学报, 2011, 37(9): 1085-1094)
[19]	Wang Xiao-Ming, Wang Shi-Tong. Theoretical analysis for the optimization problem of support vector data description. Journal of Software, 2011, 22(7): 1551-1560(王晓明, 王士同. 支撑向量数据域描述优化问题最优解理论分析. 软件学报, 2011, 22(7): 1551-1560)
[20]	Duan L X, Xu D, Tsang I W H. Domain adaptation from multiple sources: a domain-dependent regularization approach. IEEE Transactions on Neural Networks and Learning Systems, 2012, 23(3): 504-518
[21]	Duan L X, Tsang I W, Xu D, Chua T S. Domain adaptation from multiple sources via auxiliary classifiers. In: Proceedings of the 26th Annual International Conference on Machine Learning. New York, NY, USA: ACM, 2009. 289-296
[22]	Li S S, Zong C Q. Multi-domain adaptation for sentiment classification: using multiple classifier combining methods. In: Proceedings of the 2008 International Conference on Natural Language Processing and Knowledge Engineering. Beijing, China: IEEE, 2008. 1-8
[23]	Yang J, Yan R, Hauptmann A G. Cross-domain video concept detection using adaptive SVMs. In: Proceedings of the 15th International Conference on Multimedia. New York, NY: ACM, 2007. 188-197
[24]	Vapnik V N. Statistical Learning Theory. New York: John Wiley and Sons, 1998
[25]	Hofmann T, Schölkopf B, Smola A J. Kernel methods in machine learning. Annals of Statistics, 2007, 36(3): 1171-1220
[26]	Kim J, Scott C D. L2 kernel classification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(10): 1822-1831
[27]	Schölkopf B, Herbrich R, Smola A J. A generalized representer theorem. In: Proceedings of the 14th Annual Conference on Computational Learning Theory and 5th European Conference on Computational Learning Theory (COLT' 2001). London, UK: Springer-Verlag, 2001. 416-426
[28]	Dai Y H, Chen H C, Peng T. Cost-sensitive support vector machine based on weighted attribute. In: Proceedings of the 2009 International Forum on Information Technology and Applications. Washington, DC: IEEE, 2009: 690-692
[29]	Joshi M V. On evaluating performance of classifiers for rare classes. In: Proceedings of the 2nd IEEE International Conference on Data Mining. Maebashi City, Japan, 2002. 641-644