A Prior Model of Structural SVMs for Domain Adaptation

ETRI Journal

关键词: PRIOR model for structural SVMs;    structural SVMs;    Domain adaptation;   

Others  :  1186051
DOI  :  10.4218/etrij.11.0110.0571

【 摘 要 】

In this paper, we study the problem of domain adaptation for structural support vector machines (SVMs). We consider a number of domain adaptation approaches for structural SVMs and evaluate them on named entity recognition, part-of-speech tagging, and sentiment classification problems. Finally, we show that a prior model for structural SVMs outperforms other domain adaptation approaches in most cases. Moreover, the training time for this prior model is reduced compared to other domain adaptation methods with improvements in performance.

   

【 参考文献 】

• [1]B. Taskar, C. Guestrin, and D. Koller, "Max Margin Markov Networks," Proc. NIPS, vol. 16, 2004.
• [2]I. Tsochantaridis et al., "Support Vector Machine Learning for Interdependent and Structured Output Spaces," Proc. ICML, 2004.
• [3]Ben Taskar et al., "Max-Margin Parsing," Proc. EMNLP, 2004.
• [4]C. Lee and M. Jang, "Fast Training of Structured SVM Using Fixed-Threshold Sequential Minimal Optimization," ETRI J., vol. 31, no. 2, Apr. 2009, pp. 121-128.
• [5]C. Lee and M. Jang, "A Modified Fixed-Threshold SMO for 1-Slack Structural SVMs," ETRI J., vol. 32, no. 1, Feb. 2010, pp. 120-128.
• [6]C. Lee, S. Lim, and M. Jang, "Large-Margin Training of Dependency Parsers Using Pegasos Algorithm," ETRI J., vol. 32, no. 3, June 2010, pp. 486-489.
• [7]C.N. Yu et al., "Support Vector Training of Protein Alignment Models," Proc. RECOMB, 2007.
• [8]Y. Yue et al., "A Support Vector Method for Optimization Average Precision," Proc. SIGIR, 2007, pp. 271-278.
• [9]C.H. Teo et al., "A Scalable Modular Convex Solver for Regularized Risk Minimization," Proc. KDD, 2007, pp. 727-736.
• [10]T. Joachims, T. Finley, and C.N. Yu, "Cutting-Plane Training of Structural SVMs," MLJ, vol. 77, no. 1, 2008, pp. 27-59.
• [11]H. Daumé III and D. Marcu, "Domain Adaptation for Statistical Classifiers," J. Artificial Intell. Research, vol. 26, 2006, pp. 101-126.
• [12]C. Chelba and A. Acero, "Adaptation of Maximum Entropy Capitalizer: Little Data Can Help a Lot," Comput. Speech Language, vol. 20, no. 4, 2006, pp. 382-399.
• [13]H. Daumé III, "Frustratingly Easy Domain Adaptation," Proc. ACL, 2007, 2010, pp. 256-263.
• [14]J. Yang et al., "Cross-Domain Video Concept Detection Using Adaptive SVMs," Proc. 15th Int. Conf. Multimedia, 2007, pp. 188-197.
• [15]PennBioIE Corpus. http://bioie.ldc.upenn.edu/publications/latest_release/data
• [16]J. Jiang and C. Zhai, "Instance Weighting for Domain Adaptation in NLP," Proc. ACL, 2007, pp. 264-271.
• [17]Multidomain Sentiment Dataset. http://www.cs.jhu.edu/~mdredze/datasets/sentiment/
• [18]J. Blitzer, M. Dredze, and F. Pereira, "Biographies, Bollywood, Boom-boxes and Blenders: Domain Adaptation for Sentiment Classification. Association of Computational Linguistics," Proc. ACL, 2007, pp. 440-447.