文献类型：期刊文献

英文题名：Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E-Nose Systems

作者：Ma, Zhiyuan[1];Luo, Guangchun[1];Qin, Ke[1];Wang, Nan[2];Niu, Weina[1]

机构：[1]Univ Elect & Technol China, 2006 Xiyuan Ave, Chengdu 611731, Sichuan, Peoples R China;[2]East China Univ Sci & Technol, 130 Meilong Rd, Shanghai 200237, Peoples R China

年份：2018

卷号：2018

外文期刊名：WIRELESS COMMUNICATIONS & MOBILE COMPUTING

收录：;EI(收录号：20233114456399);WOS:【SCI-EXPANDED(收录号:WOS:000424819600001)】；

基金：This work was supported by the Ministry of Science and Technology Department Foundation of Sichuan Province, under Grants nos. 2016GZ0075 and 2016GZ0077, and partially supported by the Science and Technology Innovation Seed Project of Sichuan Province under Grant no. 2017RZ0008. The work of N. Wang was supported by the National Natural Science Foundation of China (NSFC) under Grant 61604054. The authors would like to thank Pei Yang from South China University of Technology and Zhipeng Cai from Georgia State University for their suggestions on refining the ideas in the paper. Additionally, the authors would like to thank Lei Zhang from Chongqing University for his instructions on implementing DAELM.

语种：英文

外文关键词：E-learning - Electronic nose - Knowledge acquisition - Online systems

摘要：Machine learning approaches have been widely used to tackle the problem of sensor array drift in E-Nose systems. However, labeled data are rare in practice, which makes supervised learning methods hard to be applied. Meanwhile, current solutions require updating the analytical model in an offline manner, which hampers their uses for online scenarios. In this paper, we extended Target Domain Adaptation Extreme Learning Machine (DAELM_T) to achieve high accuracy with less labeled samples by proposing a Weighted Domain Transfer Extreme Learning Machine, which uses clustering information as prior knowledge to help select proper labeled samples and calculate sensitive matrix for weighted learning. Furthermore, we converted DAELM T and the proposed method into their online learning versions under which scenario the labeled data are selected beforehand. Experimental results show that, for batch learning version, the proposed method uses around 20% less labeled samples while achieving approximately equivalent or better accuracy. As for the online versions, the methods maintain almost the same accuracies as their offline counterparts do, but the time cost remains around a constant value while that of offline versions grows with the number of samples.