计算机科学 ›› 2016, Vol. 43 ›› Issue (12): 173-178.doi: 10.11896/j.issn.1002-137X.2016.12.031
徐树良,王俊红
XU Shu-liang and WANG Jun-hong
摘要： 数据流挖掘已经成为数据挖掘领域一个热门的研究方向,由于数据流中概念漂移现象的存在,使得传统的分类算法无法直接应用于数据流中。为了能有效地应对数据流中的概念漂移,提出了一种基于Kappa系数的数据流分类算法。该算法采用集成式分类技术,以Kappa系数度量系统的分类性能,根据Kappa系数来动态地调整分类器,当发生概念漂移时,系统能利用已有的知识很快删除不符合要求的分类器来适应新概念。实验结果表明,相对于实验中参与比较的BWE,AE和AWE算法,该算法不但具有较好的分类性能,而且在一定程度上能较为有效地降低时间开销。
| [1] Bifet A,Holmes G,Pfahringer B.Leveraging Bagging for Evolving Data Streams [M]∥ Machine Learning and Knowledge Discovery in Databases.Springer Berlin Heidelberg,2010:135-150 [2] Lemaire V,Salperwyck C,Bondu A.A survey on SupervisedClassification on Data Streams [J].Business Intelligence,2015,205:88-125 [3] Sharker A,Hullermeier E.IBLStreams:a system for instance-based classification and regression on data streams [J].Evolving System,2012,3(4):235-249 [4] Gama J,Kosina P.Recurrent concepts in data streams classification [J].Knowledge and Information Systems,2014,40(3):489-507 [5] Bifet A,Gavalda R.Adaptive Learning from Evolving Data Str-eams [C]∥Proceedings of 8th International Symposium on Intelligent Data Analysis.Heidelberg:Springer,2009:249-260 [6] Wu Xin-dong,Li Pei-pei,Hu Xue-gang.Learning from conceptdrifting data streams with unlabeled data [J].Neurocomputing,2012,92(3):145-155 [7] Gama J,Sebastiao R,Rodrigues P P.On evaluating streamlearning algorithms [J].Machine Learning,2013,90:317-346 [8] Brzezinski D,Stefanowski J.Prequential AUC for Classifier E-valuation and Drift Detection in Evolving Data Streams [C]∥Third International Workshop NFMCP 2014 Held in Conjunction with ECML(PKDD 2014).Heidelberg:Springer,2015:87-101 [9] Rutkowski L,Pietruczuk L,Duda P,et al.Decision trees formining data streams based on the McDiarmid’s bound [J].IEEE Transactions on Knowledge and Data Engineering,2013,25(6):1272-1279 [10] Domingos P,Hulten G.Mining High- Speed Data Streams [C]∥Proceedings of the Sixth ACM SIGKDD International Confe-rence on Knowledge Discovery and Data Mining.New York:ACM,2000:71-80 [11] Magdalena.Batch Weighted Ensemble for Mining Data Streams with Concept Drift [C]∥9th International Symposium(ISMIS 2011).Heidelberg:Springer,2011:290-299 [12] Zhang Peng,Zhu Xing-quan,Shi Yong,et al.An Aggregate Ensemble for Mining Concept Drifting Data Streams with Noise [C]∥13th Pacific-Asia Conference,PAKDD 2009.Heidelberg:Springer,2009:1021-1029 [13] Wang Hai-xun,Fan Wei,Yu P S,et al.Mining Concept-Drifting Data Streams Using Ensemble Classifiers [C]∥Proceedings of the Ninth ACM SIGKDD International Conference on Know-ledge Discovery and Data Mining.New York:ACM,2003:226-235 [14] Wozniak M,Kasprzak A,Cal P.Weighted Aging Classifier Ensemble for the Incremental Drifted Data Streams [C]∥10th International Conference,FQAS 2013.Heidelberg:Springer,2013:579-588 [15] Zhou Ji-xiang,Mao Shi-song.Statistical Methods for QualityManagement [M].Beijing:China Statics Press,2008:433-440(in Chinese) 周纪芗,茆诗松.质量管理统计方法[M].北京:中国统计出版社,2008:433-440 [16] Wang Tao,Liu Ming-ju,Li De-ming.A Sstrong Chernoff Bounds Derived from Equitable Colorings of Graphs[J].Journal of Mathematics,2014,34(6):1015-1024 [17] Zliobaite I,Bifet A,Read J,et al.Evaluation methods and decision theory for classification of streaming data with temporal dependence [J].Machine Learning,2015,98(3):455-482 [18] Zhang Chen-guang,Zhang Yan.Semi-Supervised Learning [M].Beijing:China Agriculture Scientech Press,2013:31-33(in Chinese) 张晨光,张燕.半监督学习[M].北京:中国农业科学技术出版社,2013:31-33 [19] Li Pei-pei,Wu Xin-dong,Hu Xue-gang,et al.Learning concept-drifting data streams with random ensemble decision trees [J].Neurocomputing,2015,166(c):68-83 [20] Bofet A,Holmes G,Kirkby R,et al.MOA:Massive Online Analysis [J].The Journal of Machine Learning Research,2010,11(2):1601-1604 |
| No related articles found! |
|
||
首页