A Class Incremental Extreme Learning Machine for Activity Recognition

ISCAS OpenIR

	A Class Incremental Extreme Learning Machine for Activity Recognition
	Zhao, Zhongtang; Chen, Zhenyu; Chen, Yiqiang; Wang, Shuangquan; Wang, Hongan
	2014
发表期刊	COGNITIVE COMPUTATION
ISSN	1866-9956
卷号	6 期号:3 页码:423-431
摘要	Automatic activity recognition is an important problem in cognitive systems. Mobile phone-based activity recognition is an attractive research topic because it is unobtrusive. There are many activity recognition models that can infer a user's activity from sensor data. However, most of them lack class incremental learning abilities. That is, the trained models can only recognize activities that were included in the training phase, and new activities cannot be added in a follow-up phase. We propose a class incremental extreme learning machine (CIELM). It (1) builds an activity recognition model from labeled samples using an extreme learning machine algorithm without iterations; (2) adds new output nodes that correspond to new activities; and (3) only requires labeled samples of new activities and not previously used training data. We have tested the method using activity data. Our results demonstrated that the CIELM algorithm is stable and can achieve a similar recognition accuracy to the batch learning method.; Automatic activity recognition is an important problem in cognitive systems. Mobile phone-based activity recognition is an attractive research topic because it is unobtrusive. There are many activity recognition models that can infer a user's activity from sensor data. However, most of them lack class incremental learning abilities. That is, the trained models can only recognize activities that were included in the training phase, and new activities cannot be added in a follow-up phase. We propose a class incremental extreme learning machine (CIELM). It (1) builds an activity recognition model from labeled samples using an extreme learning machine algorithm without iterations; (2) adds new output nodes that correspond to new activities; and (3) only requires labeled samples of new activities and not previously used training data. We have tested the method using activity data. Our results demonstrated that the CIELM algorithm is stable and can achieve a similar recognition accuracy to the batch learning method.
收录类别	SCI
关键词	Extreme Learning Machine Incremental Learning Activity Recognition Mobile Device
部门归属	[Zhao, Zhongtang] Zhengzhou Inst Aeronaut Ind Management, Zhengzhou 450015, Peoples R China. [Zhao, Zhongtang; Chen, Zhenyu; Chen, Yiqiang; Wang, Shuangquan] Chinese Acad Sci, Pervas Comp Ctr, Inst Comp Technol, Beijing 100190, Peoples R China. [Wang, Hongan] Chinese Acad Sci, Inst Software, Beijing 100190, Peoples R China.
语种	英语
WOS记录号	WOS:000341593600012
引用统计
内容类型	期刊论文
URI标识	http://ir.iscas.ac.cn/handle/311060/16826
专题	中国科学院软件研究所
推荐引用方式 GB/T 7714	Zhao, Zhongtang,Chen, Zhenyu,Chen, Yiqiang,et al. A Class Incremental Extreme Learning Machine for Activity Recognition[J]. COGNITIVE COMPUTATION,2014,6(3):423-431.
APA	Zhao, Zhongtang,Chen, Zhenyu,Chen, Yiqiang,Wang, Shuangquan,&Wang, Hongan.(2014).A Class Incremental Extreme Learning Machine for Activity Recognition.COGNITIVE COMPUTATION,6(3),423-431.
MLA	Zhao, Zhongtang,et al."A Class Incremental Extreme Learning Machine for Activity Recognition".COGNITIVE COMPUTATION 6.3(2014):423-431.