全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

产生式与判别式线性混合分类器

, PP. 865-873

Keywords: 分类,产生式分类器,判别式分类器,遗传算法

Full-Text   Cite this paper   Add to My Lib

Abstract:

产生式方法和判别式方法是解决分类问题的两种不同框架,具有各自的优势。为利用两种方法各自的优势,文中提出一种产生式与判别式线性混合分类模型,并设计一种基于遗传算法的产生式与判别式线性混合分类模型的学习算法。该算法将线性混合分类器混合参数的学习看作一个最优化问题,以两个基分类器对每个训练数据的后验概率值为数据依据,用遗传算法找出线性混合分类器混合参数的最优值。实验结果表明,在大多数数据集上,产生式与判别式线性混合分类器的分类准确率优于或近似于它的两个基分类器中的优者。

References

[1]  Rubinstein Y D,Hastie T.Discriminative vs.Informative Learning // Proc of the 3rd International Conference on Knowledge Discovery and Data Mining.New Beach,USA,1997: 49-53
[2]  Ng A Y,Jordan M I.On Discriminative vs.Generative Classifiers: A Comparison of Logistic Regression and Nave Bayes // Becker S,Thrun S,Obermayer K,eds.Advances in Neural Information Processing Systems.Cambridge,USA: MIT Press,2002,XV: 841-848
[3]  Bishop C M,Lasserre J.Generative or Discriminative? Getting the Best of Both Worlds // Proc of the 8th World Meeting on Bayesian Statistics.Alicante,Spain,2007: 3-24
[4]  Xue J H,Titterington D M.On the Generative-Discriminative Tradeoff Approach: Interpretation,Asymptotic Efficiency and Classification Performance.Computational Statistic and Data Analysis,2010,54(2): 438-451
[5]  Xue J H,Titterington D M.Joint Discriminative-Generative Modeling Based on Statistical Tests for Classification.Pattern Recognition Letters,2010,31(9): 1048-1055
[6]  Raina R,Shen Y,Ng A Y,et al.Classification with Hybrid Generative/Discriminative Models // Thrun S,Saul L K,Schlkopf B,eds.Advances in Neural Information Processing Systems.Cambridge,USA: MIT Press,2003,XVI: 545-552
[7]  Fujino A,Ueda N,Saito K.A Hybrid Generative/Discriminative Approach to Text Classification with Additional Information.Information Processing and Management,2007,43(2): 379-392
[8]  Jaakkola T S,Haussler D.Exploiting Generative Models in Discriminative Classifiers // Kearns M J,Solla S A,Cohn D A,eds.Advances in Neural Information Processing Systems.Cambridge,USA: MIT Press,1998,II: 487-493
[9]  Tu Z W.Learning Generative Models via Discriminative Approaches // Proc of the IEEE Conference on Computer Vision and Pattern Recognition.Minneapolis,USA,2007: 1-8
[10]  Rubinstein Y D.Discriminative vs.Informative Learning.Ph.D Dissertation.Stanford,USA: Stanford University,1998
[11]  Bouchard G,Triggs B.The Trade-off between Generative and Discriminative Classifiers // Proc of the 16th International Symposium on Computational Statistics.Prague,The Czech Republic,2004: 721-728
[12]  McCallum A,Pal C,Druck G,et al.Multi-Conditional Learning: Generative/Discriminative Training for Clustering and Classification // Proc of the 20th International Conference on Artificial Intelligence.Boston,USA,2006: 433-439
[13]  Lasserre J A,Bishop C M,Minka T P.Principled Hybrid of Generative and Discriminative Models // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Miami,USA,2006: 87-94
[14]  Druck G,Pal C,McCallum A.Semi-Supervised Classification with Hybrid Generative/Discriminative Methods // Proc of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.San Jose,USA,2007: 280-289
[15]  Duda R,Hart P.Pattern Classification and Scene Analysis.New York,USA: John Wiley Sons,1973
[16]  Friedman N,Geiger D,Goldszmidt M.Bayesian Network Classifiers.Machine Learning,1997,29 (2/3): 131-163
[17]  Jing Y,Pavlovic V,Rehg J M.Boosted Bayesian Network Classifiers.Machine Learning,2008,73(1): 155-184
[18]  Hand D J,Mannila H,Smyth P.Principles of Data Mining.Cambridge,USA: MIT Press,2001
[19]  de Jong K.Learning with Genetic Algorithm: An Overview.Machine Learning,1988,3(2/3): 121-138
[20]  Witten I H,Frank E.Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations.Seattle,USA: Morgan Kaufmann Publishers,2000

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133