摘要: 由于传统的概念漂移检测研究主要针对单标签数据流,对现实中常见的多标签数据流却缺乏足够的关注,多标签数据流概念漂移检测问题有待进一步的研究。因此,通过分析多标签数据流中存在的特殊依赖关系,提出了一种基于概率相关性的多标签数据流概念漂移检测算法。其基本思想是从概念漂移的产生原因出发,利用概率相关性近似描述数据分布来监测新旧数据分布变化,判断概念漂移是否发生。实验结果表明,提出的算法能够比较快速、准确地检测到概念漂移,并在多标签概念漂移数据流分类问题上取得了预期的学习效果。
[1] Qu W,Zhang Y,Zhu J P,et al.Mining Multi-label Concept-Drif-ting Streams Using Ensemble Classifiers[C]∥Proceeding of Fuzzy Systems and Knowledge Discovery.Tianjin,China,2009,5:275-279 [2] Cheng W,Hüllermeier E,Dembczynski K J.Bayes optimal multilabel classification via probabilistic classifier chains[C]∥Proceedings of the 27th international conference on machine lear-ning.Haifa,Israel,2010:279-286 [3] Xioufis E S,Spiliopoulou M,Tsoumakas G,et al.Dealing withconcept drift and class imbalance in multi-label stream classification[C]∥Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence.Barcelona,Spain,2011,2:1583-1588 [4] Kong X,Yu P S.An ensemble-based approach to fast classification of multi-label data streams[C]∥Proceeding of 7th the International Conference on Collaborative Computing:Networking,Applications and Worksharing(CollaborateCom).Orlando,Florida,USA,2011:95-104 [5] Read J,Bifet A,Holmes G,et al.Scalable and efficient multi-label classification for evolving data streams[J].Machine Lear-ning,2012,88(1/2):243-272 [6] Cheng W,Hüllermeier E.A Simple Instance-Based Approach to Multilabel Classification Using the Mallows Model[C]∥ECML/PKKD Workshop on Learning from Multi-label Data.Bled,Slovenia,2009:28-38 [7] Hüllermeier E,Fürnkranz J,Cheng W,et al.Label ranking by learning pairwise preferences[J].Artificial Intelligence,2008,172(16):1897-1916 [8] Bifet A,Gavalda R.Learning from Time-Changing Data with Adaptive Windowing[C]∥Proceeding of the SIAM International Conference on Data Mining.Minneapolis,Minnesota,USA,2007,7:443-448 [9] Kelly M,Hand D,Adams N.The impact of changing populations on classifier performance[C]∥Proceedings of the Fifth International Conference on Knowledge Discovery and Data Mining.San Diego,CA,USA,1999:367-371 [10] Read J,Pfahringer B,Holmes G.Generating synthetic multi-label data streams[C]∥ECML/PKKD Workshop on Learning from Multi-label Data.Bled,Slovenia,2009:69-84 [11] Domingos P,Hulten G.Mining high-speed data streams[C]∥Proceedings of the sixth ACM SIGKDD international conference on Knowledge discovery and data mining.New York,USA,2000:71-80 [12] Read J,Bifet A,Holmes G,et al.Efficient multi-label classification for evolving data streams[R].University of Waikato,2010 [13] Gama J,Sebastio R,Rodrigues P P.Issues in evaluation ofstream learning algorithms[C]∥Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA,2009:329-338 |
No related articles found! |
|