基于多元区域集划分的工业数据流概念漂移检测

韩光洁; 赵腾飞; 刘立; 张帆; 徐政伟

doi:10.12263/DZXB.20221362

您当前的位置：

首页 >

文章列表页 >

基于多元区域集划分的工业数据流概念漂移检测

学术论文 | 更新时间：2025-12-08

- 基于多元区域集划分的工业数据流概念漂移检测
- Concept Drift Detection of Industrial Data Flow Based on Multivariate Region Set Partition
- 电子学报 2023年51卷第7期页码：1906-1916
- 作者机构：
  
  河海大学信息学部物联网工程学院，江苏常州 213022
- 作者简介：
  
  [ "韩光洁男，1972年8月出生于黑龙江省绥化市.现为河海大学物联网工程学院教授，博士生导师，同时也是IEEE Fellow、IET/IEE Fellow、AAIA Fellow.从事工业物联网、智慧海洋、人工智能、网络安全等方面的研究工作. E-mail: hanguangjie@gmail.com" ]
  [ "赵腾飞男，1997年9月出生于河南省周口市.现为河海大学硕士研究生.从事工业物联网故障诊断和迁移学习方面的研究工作.E-mail: zhaotengfei868@163.com" ]
  [ "刘立男，1992年5月出生于江苏省无锡市.现为河海大学物联网工程学院讲师.从事人工智能、机器学习、大数据分析方面的研究工作.E-mail: liulihhuc@gmail.com" ]
  [ "张帆女，1996年7月出生于江苏省镇江市.2019年毕业于河海大学物联网工程专业并获得学士学位.现于河海大学物联网工程学院攻读博士学位.从事工业物联网、边缘计算和机器学习方面的研究工作. E-mail: zhangfanhhuc@gmail.com" ]
  [ "徐政伟男，1994年3月出生于河南省新乡市.现于河海大学攻读计算机科学与技术专业博士学位.从事深度学习、大数据分析及辐射源个体识别等方面的研究工作. E-mail: xuzhengweihhu@outlook.com" ]
- 基金信息：
  
  国家自然科学基金(62002099);江苏省自然科学基金(BK20200184);常州市科技项目(CJ20220052);机器人学国家重点实验室联合开放基金(2022-KF-22-10)
- DOI：10.12263/DZXB.20221362
  中图分类号： TP391
- 收稿：2022-11-30，
  
  修回：2023-04-03，
  
  纸质出版：2023-07-25
- 稿件说明：
移动端阅览
韩光洁,赵腾飞,刘立等.基于多元区域集划分的工业数据流概念漂移检测[J].电子学报,2023,51(07):1906-1916.

HAN Guang-jie,ZHAO Teng-fei,LIU Li,et al.Concept Drift Detection of Industrial Data Flow Based on Multivariate Region Set Partition[J].ACTA ELECTRONICA SINICA,2023,51(07):1906-1916.
韩光洁,赵腾飞,刘立等.基于多元区域集划分的工业数据流概念漂移检测[J].电子学报,2023,51(07):1906-1916. DOI： 10.12263/DZXB.20221362.

HAN Guang-jie,ZHAO Teng-fei,LIU Li,et al.Concept Drift Detection of Industrial Data Flow Based on Multivariate Region Set Partition[J].ACTA ELECTRONICA SINICA,2023,51(07):1906-1916. DOI： 10.12263/DZXB.20221362.

摘要

为了快速适应非平稳环境中工业数据流的分布变化，需要在非结构化和噪声干扰的数据中准确、实时的完成概念漂移的检测.本文提出了一种基于多元区域集划分的工业数据流概念漂移检测算法（Concept Drift detection-Multivariate region set Partition，CDMP）.首先基于实例模糊密度进行多元区域集划分，根据划分的若干模糊分区集合，识别概念漂移发生的区域.概念漂移的持续发生会显著降低基于多元区域集构建的模型的分类性能，CDMP通过构建多元历史模型池来保留具有多样性的历史模型，以降低模型调整或再训练造成的性能损耗，同时保证概念漂移检测中准确性.CDMP在不同数据集上进行了性能测试.实验结果表明，CDMP实现了对历史模型多样性的保留和重用，能够在不同噪声水平的工业物联网环境中实现对重现型、突发型等多类型概念漂移的准确检测.

Abstract

To adapt to the rapidly changing distribution patterns generated in non-stationary industrial environments

it has become necessary to accurately and timely detect concept drift in unstructured and noisy data streams. In this study

a concept drift detection-multivariate region set partition (CDMP) algorithm for industrial data streams is proposed. The CDMP algorithm first performs multivariate region set partition based on the fuzzy density of data instances

and identifies the region in which concept drift occurs through a set of fuzzy partitions. The persistent occurrence of concept drift can significantly degrade the classification performance of models built on multivariate region sets. To address this issue

CDMP builds a historical model pool that retains diverse historical models

thus reducing the performance loss caused by model adjustment or retraining while ensuring the accuracy of concept drift detection. CDMP's performance is tested on different datasets. Experimental results show that CDMP preserves and reuses historical models with diversity

and can accurately detect different types of concept drift

including reoccurring and sudden drift

in industrial IoT environments with different levels of noise interference.

关键词

Keywords

references

YANG C E , CHEUNG Y M , DING J L , et al . Concept drift-tolerant transfer learning in dynamic environments [J]. IEEE Transactions on Neural Networks and Learning Systems , 2022 , 33 ( 8 ): 3857 - 3871 .

FRAVOLINI M L , DEL CORE G , PAPA U , et al . Data-driven schemes for robust fault detection of air data system sensors [J]. IEEE Transactions on Control Systems Technology , 2019 , 27 ( 1 ): 234 - 248 .

XU L D , HE W , LI S C . Internet of things in industries: A survey [J]. IEEE Transactions on Industrial Informatics , 2014 , 10 ( 4 ): 2233 - 2243 .

CANO A , KRAWCZYK B . Kappa updated ensemble for drifting data stream mining [J]. Machine Learning , 2020 , 109 ( 1 ): 175 - 218 .

SIDHU P , BHATIA M P S . A novel online ensemble approach to handle concept drifting data streams: Diversified dynamic weighted majority [J]. International Journal of Machine Learning and Cybernetics , 2018 , 9 ( 1 ): 37 - 61 .

ZHOU H , SHE C Y , DENG Y S , et al . Machine learning for massive industrial Internet of Things [J]. IEEE Wireless Communications , 2021 , 28 ( 4 ): 81 - 87 .

LU J , LIU A J , DONG F , et al . Learning under concept drift: A review [J]. IEEE Transactions on Knowledge and Data Engineering , 2019 , 31 ( 12 ): 2346 - 2363 .

GAMA J , ŽLIOBAITĖ I , BIFET A , et al . A survey on concept drift adaptation [J]. ACM Computing Surveys , 2014 , 46 ( 4 ): 1 - 37 .

GUO Hu-sheng , LI Hai , REN Qian-yan , et al . Concept drift type identification based on multi-sliding windows [J]. Information Sciences , 2022 , 585 : 1 - 23 .

LIU A J , ZHANG G Q , LU J . Fuzzy time windowing for gradual concept drift adaptation [C]// 2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) . Piscataway : IEEE , 2017 : 1 - 6 .

SHEN Y , DU J G , TONG J G , et al . A parallel and reverse Learn++.NSE classification algorithm [J]. IEEE Access , 2020 , 8 : 64157 - 64168 .

LI Zeng , HUANG Wen-chao , XIONG Yan , et al . Incremental learning imbalanced data streams with concept drift: The dynamic updated ensemble algorithm [J]. Knowledge-Based Systems , 2020 , 195 : 105694 .

IKONOMOVSKA E , GAMA J , DŽEROSKI S . Online tree-based ensembles and option trees for regression on evolving data streams [J]. Neurocomputing , 2015 , 150 : 458 - 470 .

GOMES H M , BIFET A , READ J , et al . Adaptive random forests for evolving data stream classification [J]. Machine Learning , 2017 , 106 ( 9 ): 1469 - 1495 .

浦慧忠 . k-means聚类分析算法在人工智能+个性化学习系统中的应用 [J]. 智能计算机与应用 , 2022 , 12 ( 8 ): 152 - 156 .

PU H Z . Application research of k-means cluster analysis algorithm in artificial intelligence + personalized learning system [J]. Intelligent Computer and Applications , 2022 , 12 ( 8 ): 152 - 156 . (in Chinese)

SAXENA A , PRASAD M , GUPTA A , et al . A review of clustering techniques and developments [J]. Neurocomputing , 2017 , 267 : 664 - 681 .

XU R , WUNSCH D . Survey of clustering algorithms [J]. IEEE Transactions on Neural Networks , 2005 , 16 ( 3 ): 645 - 678 .

RODRIGUEZ A , LAIO A . Clustering by fast search and find of density peaks [J]. Science , 2014 , 344 ( 6191 ): 1492 - 1496 .

KRAWCZYK B , MINKU L L , GAMA J , et al . Ensemble learning for data stream analysis: A survey [J]. Information Fusion , 2017 , 37 : 132 - 156 .

GHOMESHI H , GABER M M , KOVALCHUK Y . A non-canonical hybrid metaheuristic approach to adaptive data stream classification [J]. Future Generation Computer Systems , 2020 , 102 : 127 - 139 .

LIU A J , LU J , ZHANG G Q . Diverse instance-weighting ensemble based on region drift disagreement for concept drift adaptation [J]. IEEE Transactions on Neural Networks and Learning Systems , 2021 , 32 ( 1 ): 293 - 307 .

郭虎升 , 丛璐 , 高淑花 , 等 . 基于在线集成的概念漂移自适应分类方法 [J/OL]. 计算机研究与发展 : 1 - 12 . [ 2023-04-10 ]. http://kns.cnki.net/kcms/detail/11.1777.TP.20220818.163 http://kns.cnki.net/kcms/detail/11.1777.TP.20220818.163

010. html . GUO H S , CONGL, GAOS H, et al. Adaptive classification method for concept drift based on online ensemble[J/OL]. Journal of Computer Research and Development : 1 - 12 . [ 2023-04-10 ]. http://kns.cnki.net/kcms/detail/11.1777.TP.2 http://kns.cnki.net/kcms/detail/11.1777.TP.2

0220818.1639 . 010 . html . (in Chinese)

夏源 , 赵蕴龙 , 范其林 . 基于信息熵更新权重的数据流集成分类算法 [J]. 计算机科学 , 2022 , 49 ( 3 ): 92 - 98 .

XIA Y , ZHAO Y L , FAN Q L . Data stream ensemble classification algorithm based on information entropy updating weight [J]. Computer Science , 2022 , 49 ( 3 ): 92 - 98 . (in Chinese)

KRAWCZYK B , WOZNIAK M . Weighted naïve bayes classifier with forgetting for drifting data streams [C]// 2015 IEEE International Conference on Systems, Man, and Cybernetics . Piscataway : IEEE , 2015 : 2147 - 2152 .

BRZEZINSKI D , STEFANOWSKI J . Reacting to different types of concept drift: The accuracy updated ensemble algorithm [J]. IEEE Transactions on Neural Networks and Learning Systems , 2014 , 25 ( 1 ): 81 - 94 .

陆克中 , 陈超凡 , 蔡桓 , 等 . 面向概念漂移和类不平衡数据流的在线分类算法 [J]. 电子学报 , 2022 , 50 ( 3 ): 585 - 597 .

LU Ke-zhong , CHEN Chao-fan , CAI Huan , et al . Online classification algorithm for concept drift and class imbalance data stream [J]. Acta Electronica Sinica , 2022 , 50 ( 3 ): 585 - 597 . (in Chinese)

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种适用数据流概念漂移检测与适应的增量密度聚类算法

基于动态样本选择的概念漂移自适应预测方法

改进自适应模型池的在线异常检测算法

面向概念漂移和类不平衡数据流的在线分类算法