Classifier Ensemble for Imbalanced Data Stream Classification Based on Accumulated Minorities

doi:10.3969/j.issn.1006-2475.2015.03.009

Abstract

Abstract: To solve the issue of over-fitting and not making full use of current data in existing methods of balancing imbalanced data stream, a method named EAMIDS for imbalanced data stream is proposed based on accumulated positive samples. In EAMIDS, positive samples from previous training chunks are accumulated to form the AP set which is used to balance the class distributions by making use of K nearest neighbors and Over-sampling technique. The ensemble classifier will be updated according to F-Measure when the number of the available base classifiers is greater than the fixed size of the ensemble classifier. Empirical study on both SEA dataset and SPH dataset shows that the proposed EAMIDS has substantial advantage over IDSL approach and SMOTE approach in prediction accuracy.


Key words: imbalanced data streams, accumulated positive samples, ensemble classifiers, concept drift

CLC Number:

TP391

GUO Wen-feng1， WANG Yong2. Classifier Ensemble for Imbalanced Data Stream Classification Based on Accumulated Minorities[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2015.03.009.

References

［1]Valdovinos R M, Sanchez J S. Class-dependant resampling for medical applications［C]// Proceedings of the 4th International Conference on Machine Learning and Applications. 2005:351-356.
［2]Fawcett T, Provost F. Adaptive fraud detection［J]. Data Mining and Knowledge Discovery, 1997,1(3):291-316.
［3]Japkowicz N, Stephen S. The class imbalance problem: A systematic study［J]. Intelligent Data Analysis, 2002,6(5):429-449.
［4]Chawla N V, Bowyer K W, Hall L O, et al. SMOTE:Synthetic minority Over-sampling technique［J]. Journal of Artificial Intelligence Research, 2002,16:321-357.
［5]Chen Sheng, He Haibo. Sera: Selectively recursive approach towards nonstationary imbalanced stream data mining［C]// International Joint Conference on Neural Networks, 2009. 2009:522-529.
［6]Chen Sheng, He Haibo. Towards incremental learning of nonstationary imbalanced data stream: A multiple selectively recursive approach［J]. Evolving Systems, 2011,2(1):35-50.
［7]Godase A, Attar V. Classifier ensemble for imbalaced data stream classification［C]// Proceedings of the CUBE International Information Technology Conference, 2012. 2012:284-289.
［8]Gao Jing, Ding Bolin, Fan Wei, et al. Classifying data streams with skewed class distributions and concept drifts［J]. IEEE Internet Computing, 2008,12(6):37-49.
［9]Zhang Juan, Hu Xuegang, Zhang Yuhong, et al. An efficient ensemble method for classifying skewed data streams［C]// Proceedings of the 7th International Conference on Intelligent Computing: Bio-inspired Computing and Applications. 2011:144-151.
［10]欧阳震诤,罗建书,胡东敏,等. 一种不平衡数据流集成分类模型［J]. 电子学报, 2010,38(1):184-189.
［11]Street W N, Kim Yongseog. A streaming ensemble algorithm(sea) for large-scale classification［C] // Proceedings of the 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2001:377-382.
［12]Witten I H, Frank E. Data Mining: Practical Machine Learning Tools and Techniques［M]. 2nd ed. San Fransisco: Morgan Kaufmann, 2005.
［13]Van Rijsbergen C V. Information Retrieval［M]. 2nd ed. London: Butterworths, 1979.
［14]Kubat M, Holte C R, Matwin S. Learning when negative examples abound［C]// Proceedings of the 9th European Conference Machine Learning. 1997,1224:146-153.
［15]Wang Haixun, Fan Wei, Yu P S, et al. Mining concept-drifting data streams using ensemble classifiers［C]// Proceedings of the 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2003:226-235.

[1]	ZHANG Heng, JU Shi-guang. Network Data Stream Classification Based on Concept Drift Detection [J]. Computer and Modernization, 2021, 0(07): 107-114.
[2]	XIE Yan-song;ZHU Xiao-lian. Analysis of Classification Technology of Data Stream [J]. Computer and Modernization, 2010, 1(01): 32-36.