第2期 文益民，等：概念漂移数据流分类研究综述 ·101· Learning.Helsinki,Finland,1993:244-261. 1999:41-48. [23]BLACK M,HICKEY R J.Maintaining the performance of [35]KUBAT G,GAMA J,UTGOFF P.Special issue on incre- a learned classifier under concept drift[J].Intelligent Da mental learning systems capable of dealing with concept ta Analys3is,1999,3(6):453474. drift[Z].Amsterdam:Intelligent Data Analysis,2004. [24]KUN A,PETSCHE T,RIVEST R L.Learning time-var- [36]GAO J,DING B,FAN W,et al.Classifying data stream ying concepts[C]//Proceedings of the Conference Neural with skewed class distributions and concept drifts J]. Information Processing Systems.Denver,USA,1990: IEEE Transactions on Internet Computing,2008,12(6): 3749. 183-189. [37]CHEN S,HE H.SERA:selectively recursive approach to- [25]HELMBOLD D P,LONG P M.Tracking drifting concepts wards nonstationary imbalanced stream data mining[C]// by minimizing disagreements [J].Machine Learning, Proceedings of the 2009 International Joint Conference on 1994,14(1):27-45. Neural Networks.Atlanta,USA,2009:522-529. [26]BARVE R D,LONG P M.On the complexity of leaming [38]CHEN S,HE H,LI K,et al.MuSeRA:multiple selec- from drifting distributions C]//Proceedings of the Ninth tively recursive approach towards nonstationary imbalanced Annual Conference on Computational Leaming Theory. stream data mining[C]//Proceedings of the 2010 Intera- Desenzano Sul Garda,Italy,1996:122-130. tional Joint Conference on Neural Networks.Barcelona. [27]DIETTERICH T,WIDMER G,KUBAT M.Special issue Spain,2010:18. on context sensitivity and concept drift[J].Machine [39]CHEN S,HE H.Towards incremental learning of nonsta- Learning,1998,32(2):83-201. tionary imbalanced data stream:a multiple selectively re- [28]WANG H,FAN W,YU P S,et al.Mining concept-drif- cursive approach[J].Evolving Systems,2010,2(1):35- ting data streams using ensemble classifiers[C]//Proceed- 50. ings of the Ninth International Conference on Knowledge [40]LICHTENWALTER R N,CHAWLA N V.Adaptive meth- Discovery and Data Mining.Washington,DC,USA, ods for classification in arbitrarily imbalanced and drifting 2003:226-235. data streams C]//Proceedings of the 13th Pacific-Asia [29]KOLTER J Z,MALOOF M A.Dynamic weighted majori- Conference on Knowledge Discovery and Data Mining. Bangkok,Thailand,2009:53-75. ty:a new ensemble method for tracking concept drift [41]GREGORY D,ROBI P.An ensemble based incremental [C]//Proceedings of the Third IEEE International Confer- leamning framework for concept drift and class imbalance ence on Data Mining.Washington,DC,USA,2003:123- [C]//Proceedings of the 2010 International Joint Confer- 130. ence on Neural Networks.Barcelona,Spain,2010:9-17. [30]KOLTER J Z,MALOOF M A.Using additive expert en- [42]ZHANG J,HU X,ZHANG Y,et al.An efficient ensem- sembles to cope with concept drift[C]//Proceedings of the ble method for classifying skewed data streams[C]//Pro- 22nd International Conference on Machine Learning. ceedings of the 7th International Conference on Intelligent Bonn,Germany,2005:449-456. Computing.Zhengzhou,China,2011:144-151. [31]BIFET A,HOLMES G,PFAHRINGER B,et al.New en- [43 RAMAMURTHY S,BHATNAGAR R.Tracking recurrent semble methods for evolving data streams[C]//Proceed- concept drift in streaming data using ensemble classifiers ings of the 15th ACM SIGKDD International Conference on [C]//Proceedings of the 6th International Conference on Knowledge Discovery and Data Mining.Paris,France, Machine Learing and Applications.Cincinnati,USA, 2009:139-148. 2007:404409. [32]KLINKENBERG R.Using labeled and unlabeled data to [44]LI PP,WU X D,HU X G.Mining recurring concept learning drifting concepts[C]//Proceedings of the Work- drifts with limited labeled streaming data[C]//Proceed- shop Notes of the IJCAI-01 Workshop on Learning from ings of the 2th Asian Conference on Machine Learning.To- Temporal and Spatial Data.Menlo Park,USA 2001:16- kyo,Japan,2010:241-252. 24. [45 ]MASUD MM,AL-KHATEEB T M,KHAN L,et al.De- [33]KLINKENBERG R.Detection concept drift with support tecting recurring and novel classes in concept-drifting data vector machines[C]//Proceedings of the Seventeenth In- streams[C]//Proceedings of the 2011 International Con- temational Conference on Machine Learning.Stanford, ference on Data Mining.Vancouver,Canada,2011:1176- California,USA,2000:487-494. 1181. [34 LANQUILLON C.Information filtering in changing do- [46]XUE J C,WEISS G M.Quantification and semi-super- mains[C]//Proceedings of the 16th International Joint vised classification methods for handling changes in class Conference on Artificial Intelligence.Stockholm,Sweden, distribution[C]//Proceedings of the 15th ACM SIGKDD