计算机科学 ›› 2018, Vol. 45 ›› Issue (3): 189-195.doi: 10.11896/j.issn.1002-137X.2018.03.030

• 人工智能 • 上一篇    下一篇

一种基于树型贝叶斯网络的集成多标记分类算法

张志东,王志海,刘海洋,孙艳歌   

  1. 北京交通大学计算机与信息技术学院 北京100044,北京交通大学计算机与信息技术学院 北京100044,北京交通大学计算机与信息技术学院 北京100044,北京交通大学计算机与信息技术学院 北京100044
  • 出版日期:2018-03-15 发布日期:2018-11-13
  • 基金资助:
    本文受国家自然科学基金(61672086),北京市自然科学基金(4182052)资助

Ensemble Multi-label Classification Algorithm Based on Tree-Bayesian Network

ZHANG Zhi-dong, WANG Zhi-hai, LIU Hai-yang and SUN Yan-ge   

  • Online:2018-03-15 Published:2018-11-13

摘要: 在多标记分类问题中,有效地利用标记间的依赖关系是进一步提升分类器性能的主要途径之一。基于分类器链算法,利用互信息度量理论构造分类对象的类属性之间明确的多标记关系依赖模型,并依据建立的标记依赖模型将分类器链中的线性依赖拓展成树型依赖,以适应更为复杂的标记依赖关系;同时,在此基础上利用Stacking集成学习方法建立最终训练模型,提出了一种新的针对树型依赖表示模型的Stacking算法。 在多个实验数据集上的实验结果表明,与原有的Stacking集成学习相比,该算法提升了分类器的相应评价指标。

关键词: 多标记分类,标记依赖,Stacking,树型贝叶斯网络

Abstract: The performance of learning algorithm can be improved by utilizing existing label dependencies in multi-label classification.Based on the strategy of classifier chain and stacking ensemble learning,this paper built a model to explain the dependency of different labels,and extended the linear dependency into tree dependency to deal with much more complicated label relations.Compared with the original Stacking algorithm,the performance of the proposed algorithm is improved in the experiments.

Key words: Multilabel classification,Label dependency,Stacking,Tree-Bayesian network

[1] TSOUMAKAS G,KATAKIS I,VLAHAVAS I.Mining Multi-label Data[M]∥Data Mining and Knowledge Discovery Handbook.Boston:Springer,2009:667-685.
[2] ZHANG M L,ZHOU Z H.A review on multi-label learning algorithms[J].IEEE Transactions on Knowledge & Data Engineering,2014,26(8):1819-1837.
[3] UEDA N,SAITO K.Parametric mixture model for multitopictext[J].Systems and Computers in Japan,2006,37(2):56-66.
[4] TSOURAKAKIS C.Provably fast inference of latent featuresfrom networks:with applications to learning social circles and multilabel classification[C]∥Proceedings of the 24th International Conference on World Wide Web.ACM,2015:1111-1121.
[5] LUO Y,LIU T,TAO D,et al.Multiview matrix completion for multilabel image classification[J].IEEE Transactions on Image Processing,2015,24(8):2355-2368.
[6] ZHANG M L,ZHOU Z H.ML-KNN:A lazy learning approach to multi-label learning[J].Pattern Recognition,2007,40(7):2038-2048.
[7] CLARE A,KING R D.Knowledge Discovery in Multi-labelPhenotype Data[J].Lecture Notes in Computer Science,2002,2168(2168):42-53.
[8] ELISSEEFF A E,WESTON J.A Kernel Method for Multi-Labelled Classification[C]∥Advances in Neural Information Processing Systems.2002:681-687.
[9] GHAMRAWI N,MCCALLUM A.Collective multi-label classification[C]∥Proceedings of the 14th ACM International Conference on Information and Knowledge Management.ACM,2005:195-200.
[10] RNKRANZ J,LLERMEIER E,LOZAMENC,et al.Multilabel classification via calibrated label ranking[J].Machine Learning,2008,73(2):133-153.
[11] ALVARES-CHERMAN E,METZ J,MONARD M C.Incorporating label dependency into the binary relevance framework for multi-label classification[J].Expert Systems with Applications,2012,39(2):1647-1655.
[12] TSOUMAKAS G,VLAHAVAS I.Random k-Labelsets:An Ensemble Method for Multilabel Classification[C]∥ European Conference on Machine Learning.Springer,Berlin,Heidelberg,2007:406-417.
[13] READ J,PFAHRINGER B,HOLMES G,et al.Classifier chains for multi-label classification[J].Machine Learning,2011,85(3):254-269.
[14] TSOUMAKAS G,DIMOU A,SPYROMITROS E,et al.Correlation-based pruning of stacked binary relevance models for multi-label learning[C]∥Proceedings of the 1st International Workshop on Learning from Multi-Label Data.2009:101-116.
[15] DEMBCZY'SKI K,CHENG W,HLLERMEIER E.Bayes optimal multilabel classification via Probabilistic Classifier Chains[C]∥International Conference on Machine Learning.2010:279-286.
[16] ZHANG M L,ZHOU Z H.Multilabel neural networks with applications to functional genomics and text categorization[J].IEEE Transactions on Knowledge and Data Engineering,2006,18(10):1338-1351.
[17] CHENG W,HLLERMEIER E.Combining instance-basedlearning and log is ticre gression for multilabel classification[J].Machine Learning,2009,76(2-3):211-225.
[18] ALVARES-CHERMAN E,METZ J,MONARD M C.Incorporating label dependency into the binary relevance framework for multi-label classification[J].Expert Systems with Applications,2012,39(2):1647-1655.
[19] BIELZA C,LI G,LARRAAGA P.Multi-dimensional classification with Bayesian networks[J].International Journal of Approximate Reasoning,2011,52(46):705-727.
[20] FU B,WANG Z,XU G,et al.Multi-label learning based on itera-tive label propagation over graph[J].Pattern Recognition Letters,2014,42(1):85-90.
[21] DE CAMPOS L M.A Scoring Function for Learning Bayesian Networks based on Mutual Information and Conditional Independence Tests[J].Journal of Machine Learning Research,2006,7(7):2149-2187.
[22] FU B,WANG Z H.Multi-Label Classification Method Based on Tree Structure of Label Dependency[J].Pattern Recognition and Artificial Intelligence,2012,25(4):573-580.(in Chinese) 付彬,王志海.基于树型依赖结构的多标记分类算法[J].模式识别与人工智能,2012,25(4):573-580.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!