A novel random forests based class incremental learning method for activity recognition

doi:10.1016/j.patcog.2018.01.025

	A novel random forests based class incremental learning method for activity recognition
	Hu, Chunyu 1,2,3; Chen, Yiqiang 1,2,3; Hu, Lisha 1,2,3,4; Peng, Xiaohui 1,3
	2018-06-01
发表期刊	PATTERN RECOGNITION
ISSN	0031-3203
卷号	78 页码:277-290
摘要	Automatic activity recognition is an active research topic which aims to identify human activities automatically. A significant challenge is to recognize new activities effectively. In this paper, we propose an effective class incremental learning method, named Class Incremental Random Forests (CIRF), to enable existing activity recognition models to identify new activities. We design a separating axis theorem based splitting strategy to insert internal nodes and adopt Gini index or information gain to split leaves of the decision tree in the random forests (RF). With these two strategies, both inserting new nodes and splitting leaves are allowed in the incremental learning phase. We evaluate our method on three UCI public activity datasets and compare with other state-of-the-art methods. Experimental results show that the proposed incremental learning method converges to the performance of batch learning methods (RF and extremely randomized trees). Compared with other state-of-the-art methods, it is able to recognize new class data continuously with a better performance. (C) 2018 Elsevier Ltd. All rights reserved.
关键词	Class incremental learning Activity recognition Random forests
DOI	10.1016/j.patcog.2018.01.025
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2017YFB1002801] ; Natural Science Foundation of China[61572471] ; Chinese Academy of Sciences Pioneer Hundred Talents Program[Y704061000] ; Science and Technology Planning Project of Guangdong Province, China[2015B010105001]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000428490900021
出版者	ELSEVIER SCI LTD
引用统计	被引频次：76[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/2XEOYT63/5900
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Chen, Yiqiang
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 2.Univ Chinese Acad Sci, Beijing, Peoples R China 3.Beijing Key Lab Mobile Comp & Pervas Device, Beijing, Peoples R China 4.Hebei Univ Econ & Business, Shijiazhuang, Hebei, Peoples R China
推荐引用方式 GB/T 7714	Hu, Chunyu,Chen, Yiqiang,Hu, Lisha,et al. A novel random forests based class incremental learning method for activity recognition[J]. PATTERN RECOGNITION,2018,78:277-290.
APA	Hu, Chunyu,Chen, Yiqiang,Hu, Lisha,&Peng, Xiaohui.(2018).A novel random forests based class incremental learning method for activity recognition.PATTERN RECOGNITION,78,277-290.
MLA	Hu, Chunyu,et al."A novel random forests based class incremental learning method for activity recognition".PATTERN RECOGNITION 78(2018):277-290.