Fast semi-supervised self-training algorithm based on data editing

doi:10.1016/j.ins.2023.01.029

	Fast semi-supervised self-training algorithm based on data editing
	Li, Bing1,2 ; Wang, Jikui2 ; Yang, Zhengguo2 ; Yi, Jihai2 ; Nie, Feiping3,4
	2023-05
发表期刊	INFORMATION SCIENCES
卷号	626 页码:293-314
摘要	Self-training is a commonly semi-supervised learning Algorithm framework. How to select the high-confidence samples is a crucial step for algorithms based on self-training framework. To alleviate the impact of noise data, researchers have proposed many data editing methods to improve the selection quality of high-confidence samples. However, the state-of-the-art data editing methods have high time complexity, which is not less than O(n(2)), where n denotes the number of samples. To improve the training speed while ensuring the quality of the selected high-confidence samples, inspired by Ball-k-means algorithm, we propose a fast semi-supervised self-training Algorithm based on data editing (EBSA), which defines ball-cluster partition and editing to improve the quality of high-confidence samples. The time complexity of the proposed EBSA is O(t(2kn + n log n + n + k(2))) , where k denotes the number of centers, t denotes the number of iterates. k is far less than n, EBSA has linear time complexity with respect to n. A large number of experiments on 20 benchmark data sets have been carried out and the experimental results show that the proposed Algorithm not only ran faster, but also obtained better classification performance compared with the comparison algorithms. (c) 2023 Elsevier Inc. All rights reserved.
关键词	Semi-supervised learning Self-training classification Ball-k-means Data editing
DOI	10.1016/j.ins.2023.01.029
收录类别	SCIE ; EI
ISSN	0020-0255
语种	英语
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems
WOS记录号	WOS:000925337600001
出版者	ELSEVIER SCIENCE INC
EI入藏号	20230413427395
EI主题词	Classification (of information)
EI分类号	716.1 Information Theory and Signal Processing ; 723.4.2 Machine Learning ; 903.1 Information Sources and Analysis
原始文献类型	Article
EISSN	1872-6291
引用统计	被引频次：7[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/33449
专题	信息工程与人工智能学院
作者单位	1.Guizhou Univ, State Key Lab Publ Big Data, Guiyang 550025, Guizhou, Peoples R China; 2.Lanzhou Univ Finance & Econ, Coll Informat Engn, Lanzhou 730020, Gansu, Peoples R China; 3.Northwestern Polytech Univ, Sch Comp Sci, Xian 710072, Shanxi, Peoples R China; 4.Northwestern Polytech Univ, Ctr OPT IMagery Anal & Learning OPTIMAL, Xian 710072, Shanxi, Peoples R China
第一作者单位	兰州财经大学
推荐引用方式 GB/T 7714	Li, Bing,Wang, Jikui,Yang, Zhengguo,et al. Fast semi-supervised self-training algorithm based on data editing[J]. INFORMATION SCIENCES,2023,626:293-314.
APA	Li, Bing,Wang, Jikui,Yang, Zhengguo,Yi, Jihai,&Nie, Feiping.(2023).Fast semi-supervised self-training algorithm based on data editing.INFORMATION SCIENCES,626,293-314.
MLA	Li, Bing,et al."Fast semi-supervised self-training algorithm based on data editing".INFORMATION SCIENCES 626(2023):293-314.