Adaptive bagging-based dynamic ensemble selection in nonstationary environments

被引:0
|
作者
Wei, Bo [1 ,3 ]
Chen, Jiakai [1 ]
Deng, Li [2 ]
Mo, Ziyan [2 ]
Jiang, Mingfeng [1 ]
Wang, Feng [4 ]
机构
[1] Zhejiang Sci Tech Univ, Sch Comp Sci & Technol, Sch Artificial Intelligence, Hangzhou 310018, Zhejiang, Peoples R China
[2] Zhejiang Sci Tech Univ, Sch Sci, Hangzhou 310018, Zhejiang, Peoples R China
[3] Zhejiang Sci Tech Univ, Longgang Res Inst, Longgang 325000, Zhejiang, Peoples R China
[4] Wuhan Univ, Sch Comp Sci, Wuhan 430072, Hubei, Peoples R China
基金
中国国家自然科学基金;
关键词
Nonstationary environments; Concept drift; Imbalance learning; Ensemble learning; CLASSIFIER SELECTION; FRAMEWORK; IMBALANCE;
D O I
10.1016/j.eswa.2024.124860
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The classification of nonstationary data streams presents a substantial challenge mainly due to the simultaneous presence of class imbalance and concept drift. Ensemble learning is a widely acknowledged approach to address these two problems. Nevertheless, most of the existing methods establish independent strategies for each, ignoring their relationship and interactions, which prevents the corresponding strategies from performing as expected. This paper presents a novel algorithm AB-DES (Adaptive Bagging-based Dynamic Ensemble Selection), which employs bagging technique and dynamic ensemble selection system. In AB-DES, instances are stored to balance subsequent data chunks instead of relying solely on resampling methods. Then, all base classifiers are updated using a new dual adaptive sampling bagging strategy, which can help to reduce the classification inaccuracies involving by class imbalance. In the prediction phase, DES system constructs different ensemble models for different instances. The inclusion of a sliding window-based drift detector makes the above two strategies adapt to concept drift. Experimental results show that AB-DES achieves higher classification accuracy and exhibits enhanced robustness compared to 8 state-of-the-art methods on 10 synthetic datasets and 5 real world datasets with different types of concept drift.
引用
收藏
页数:13
相关论文
共 50 条
  • [31] NLOS identification and mitigation in UWB positioning with bagging-based ensembled classifiers
    V Ch Sekhar Rao Rayavarapu
    Arunanshu Mahapatro
    [J]. Annals of Telecommunications, 2022, 77 : 267 - 280
  • [32] An ensemble approach for incremental learning in nonstationary environments
    Muhlbaier, Michael D.
    Polikar, Robi
    [J]. MULTIPLE CLASSIFIER SYSTEMS, PROCEEDINGS, 2007, 4472 : 490 - +
  • [33] Intrusion Detection System using Bagging Ensemble Selection
    Sreenath, M.
    Udhayan, J.
    [J]. 2015 IEEE INTERNATIONAL CONFERENCE ON ENGINEERING AND TECHNOLOGY (ICETECH), 2015, : 4 - 7
  • [34] Bagging-Based Active Learning Model for Named Entity Recognition with Distant Supervision
    Lee, Sunghee
    Song, Yeongkil
    Choi, Maengsik
    Kim, Harksoo
    [J]. 2016 INTERNATIONAL CONFERENCE ON BIG DATA AND SMART COMPUTING (BIGCOMP), 2016, : 321 - 324
  • [35] Bagging.LMS: A bagging-based linear fusion with least-mean-square error update for regression
    Wu, Yunfeng
    Wang, Cong
    Ng, S. C.
    [J]. TENCON 2006 - 2006 IEEE REGION 10 CONFERENCE, VOLS 1-4, 2006, : 207 - +
  • [36] Adaptive Selection of Classifier Ensemble Based on GMDH
    Xiao, Jin
    He, Changzheng
    [J]. 2008 INTERNATIONAL SEMINAR ON FUTURE INFORMATION TECHNOLOGY AND MANAGEMENT ENGINEERING, PROCEEDINGS, 2008, : 61 - 64
  • [37] HRRP recognition for polarization radar based on Bagging-SVM dynamic ensemble
    Zhang, Yu-Xi
    Wang, Xiao-Dan
    Yao, Xu
    Bi, Kai
    [J]. Zhang, Y.-X. (zhyuxibj@163.com), 1600, Chinese Institute of Electronics (34): : 1366 - 1371
  • [38] Bagging based ensemble transfer learning
    Liu, Xiaobo
    Wang, Guangjun
    Cai, Zhihua
    Zhang, Harry
    [J]. JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING, 2016, 7 (01) : 29 - 36
  • [39] Bagging based ensemble transfer learning
    Xiaobo Liu
    Guangjun Wang
    Zhihua Cai
    Harry Zhang
    [J]. Journal of Ambient Intelligence and Humanized Computing, 2016, 7 : 29 - 36
  • [40] Landslide Susceptibility Modeling Based on GIS and Novel Bagging-Based Kernel Logistic Regression
    Chen, Wei
    Shahabi, Himan
    Zhang, Shuai
    Khosravi, Khabat
    Shirzadi, Ataollah
    Chapi, Kamran
    Binh Thai Pham
    Zhang, Tingyu
    Zhang, Lingyu
    Chai, Huichan
    Ma, Jianquan
    Chen, Yingtao
    Wang, Xiaojing
    Li, Renwei
    Bin Ahmad, Baharin
    [J]. APPLIED SCIENCES-BASEL, 2018, 8 (12):