Disagreement-based class incremental random forest for sensor-based activity recognition

doi:10.1016/j.knosys.2021.108044

	Disagreement-based class incremental random forest for sensor-based activity recognition
	Hu, Chunyu 1,2,3; Chen, Yiqiang 3,4,5; Hu, Lisha 6; Yu, Han 7; Lu, Dianjie 2,8
	2022-03-05
发表期刊	KNOWLEDGE-BASED SYSTEMS
ISSN	0950-7051
卷号	239 页码:15
摘要	Activity recognition plays a key role in many fields, such as health monitoring and elderly care. Handling changes in user habits is a significant technical challenge in activity recognition. Ideally, a model should adapt to newly emerging classes and concept drift dynamically. This paper proposes a novel semi-supervised class incremental learning method, namely, disagreement-based class incremental random forest (Di-CIRF). The proposed model can detect newly emerging classes and update a previously established activity recognition model through streaming data. First, it is necessary to identify novel candidates by employing the disagreement-based confidence voting mechanism and minimum bounding box (MBB)-based separation detection to annotate newly emerging data accurately. Then, the coarse coding-based cohesion detection strategy is adopted to filter out the true novelty instances. This paper also proposes the iterative MBB-based splitting strategy and the pseudo-instance generation mechanism in Di-CIRF for updating the activity model without retaining the trained data. According to experimental results on four public activity recognition datasets, Di-CIRF outperforms the state-of-the-art methods.
关键词	Class incremental learning Activity recognition Random forest Semi-supervised learning
DOI	10.1016/j.knosys.2021.108044
收录类别	SCI
语种	英语
资助项目	Key-Area Research and Development Program of Guangdong Province, China[2019B010109001] ; National Natural Science Foundation of China[62002187] ; National Natural Science Foundation of China[61972237] ; Nat-ural Science Foundation of Hebei Province, China[F2019207061] ; Nanyang Technological University, Singapore, through the Nanyang Assistant Professorship (NAP)
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000806866700004
出版者	ELSEVIER
引用统计	被引频次：4[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/19603
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Hu, Chunyu; Hu, Lisha
作者单位	1.Qilu Univ Technol, Sch Comp Sci & Technol, Shandong Acad Sci, Jinan, Peoples R China 2.Shandong Prov Key Lab Distributed Comp Software N, Jinan, Peoples R China 3.Shandong Acad Intelligent Comp Technol, Jinan, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 5.Univ Chinese Acad Sci, Beijing, Peoples R China 6.Hebei Univ Econ & Business, Inst Informat Technol, Shijiazhuang, Hebei, Peoples R China 7.Nanyang Technol Univ, Sch Comp Sci & Engn, Singapore, Singapore 8.Shandong Normal Univ, Sch Informat Sci & Engn, Jinan, Peoples R China
推荐引用方式 GB/T 7714	Hu, Chunyu,Chen, Yiqiang,Hu, Lisha,et al. Disagreement-based class incremental random forest for sensor-based activity recognition[J]. KNOWLEDGE-BASED SYSTEMS,2022,239:15.
APA	Hu, Chunyu,Chen, Yiqiang,Hu, Lisha,Yu, Han,&Lu, Dianjie.(2022).Disagreement-based class incremental random forest for sensor-based activity recognition.KNOWLEDGE-BASED SYSTEMS,239,15.
MLA	Hu, Chunyu,et al."Disagreement-based class incremental random forest for sensor-based activity recognition".KNOWLEDGE-BASED SYSTEMS 239(2022):15.