Message Board

Respected readers, authors and reviewers, you can add comments to this page on any questions about the contribution, review,        editing and publication of this journal. We will give you an answer as soon as possible. Thank you for your support!

Name
E-mail
Phone
Title
Content
Verification Code
Turn off MathJax
Article Contents

YU Xin, ZHENG Zhaobao, LI Linyi. Research on Oblique Factor Model for Selecting Training Samples[J]. Geomatics and Information Science of Wuhan University. doi: 10.13203/j.whugis20200631
Citation: YU Xin, ZHENG Zhaobao, LI Linyi. Research on Oblique Factor Model for Selecting Training Samples[J]. Geomatics and Information Science of Wuhan University. doi: 10.13203/j.whugis20200631

Research on Oblique Factor Model for Selecting Training Samples

doi: 10.13203/j.whugis20200631
Funds:

the National Key Research and Development Program of China (No. 2018YFC0407804).

  • Received Date: 2020-12-10
  • Objectives: Researchers notice that the quality of training samples will impact the effective of training phase and then further will have an influence on the overall classification accuracy in the testing phase. In fact, representativeness or typicalness of training samples is able to reflect the quality of training samples in a way. Especially for the currently popular deep learning methods, it has needed thousands or millions of training samples. Therefore, how to reduce the number of training samples for deep learning method becomes a very important problem. In another hand, from the actual application angle, it is also very expensive. Therefore, we research one method of reducing the training samples as less as possible based on the representativeness or typicalness of training samples. Method: selection of training samples based on oblique factor model is proposed and it relaxes the independent condition among common factors in the orthogonal factor model, which is able to better describe the real world. Results: Experimental results show the proposed method is feasible and effective and it is able to select more representative training samples than the method of selection of training samples based on orthogonal factor model and achieve better performance in the overall classification precision and stability. Experimental results show that selection of training samples based on oblique factor model outperforms selection of training samples based on orthogonal factor model. And the distribution of selected samples becomes more decentralized and reasonable and the overall classification accuracy averagely improves about 3%. Conclusions: the proposed method, not only supports how to optimize capturing data in the theory, but also is able to guide how to effectively capture data in the actual application.
  • [1] Adeli Ehsan, Li Xiaorui, Kwon Dongjin, Zhang Yong, Pohl Kilian M. Logistic Regression Confined by Cardinality-Constrained Sample and Feature Selection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(7):1713-1728.
    [2] Arnold Roger B., Wang Luke, Lopez Talle, James Sophie, Blute Nicole. Updating Lead and Copper Rule SampleSite Selection:Best Practices From an Innovative Pilot Program[J]. Journal of American Water Works Association, 2020, 112(4):22-31.
    [3] Au Jessie, Youngentob Kara N, Foley William J, Moore Ben D, Fearn Tom. Sample selection, calibration and validation of models developed from a large dataset of near infrared spectra of tree leaves[J]. Journal of Near Infrared Spectroscopy, 2020. (Article in Press).
    [4] Bellver Miriam, Salvador Amaia, Torres Jordi, Giro-i-Nieto Xavier. Mask-guided sample selection for semi-supervised instance segmentation[J]. Multimedia Tools and Applications, 2020. (Article in Press).
    [5] da Silva, Marcus Vinicius Brito, de Carvalho, André Augusto Pacheco, Jacobs Arthur Selle, Pfitscher Ricardo José, Granville Lisandro Zambenedetti. Sample Selection Search to Predict Elephant Flows in IXP Programmable Networks[J]. Advances in Intelligent Systems and Computing, 2020, 1151:357-368.
    [6] Fernández Mariela, García Jesús E., Gholizadeh Ramin, González-López Verónica A. Sample selection procedure in daily trading volume processes[J]. Mathematical Methods in the Applied Sciences, 2020, 43(13):7537-7549.
    [7] He Kaixun, Wang Kai, Yan Yayun. Active training sample selection and updating strategy for near-infrared model with an industrial application[J]. Chinese Journal of Chemical Engineering, 2019, 27(11):2749-2758.
    [8] Kral, Jan, Gotthans Tomas, Marsalek Roman, Harvanek Michal, Rupp Markus. On feedback sample selection methods allowing lightweight digital predistorter adaptation[J]. IEEE Transactions on Circuits and Systems I:Regular Papers, 2020, 67(6):1976-1988.
    [9] Li Huiyong, Bao Weiwei, Hu Jinfeng, Xie Julan, Liu Ruixin. A training samples selection method based on system identification for STAP[J]. Signal Processing, 2018, 142:119-124.
    [10] Liu Jing, Zhu A-Xing, Rossiter David, Du Fei, Burt James. A trustworthiness indicator to select sample points for the individual predictive soil mapping method (iPSM)[J]. Geoderma, 2020, 373.
    [11] Liu Xueqi, Zhu A-Xing, Yang Lin, Pei Tao, Liu Junzhi, Zeng Canying, Wang Desheng. A graded proportion method of training sample selection for updating conventional soil maps[J]. Geoderma, 2020, 357. (Open Access)
    [12] Lu Qikai, Ma Yong, Xia Gui-Song. Active learning for training sample selection in remote sensing image classification using spatial information[J]. Remote Sensing Letters, 2017, 8(12):1210-1219.
    [13] Lu Wenbo, Ma Chaoqun, Li Peikun. Research on Sample Selection of Urban Rail Transit Passenger Flow Forecasting Based on SCBP Algorithm[J]. IEEE Access, 2020, 8:89425-89438.
    [14] Lu Yang, Ma Xiaolei, Lu Yinan. A cluster-based sample selection strategy for biological event extraction[C].//Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering, 2019, p 72-77.
    [15] Ma Jing, Hong Dezhi, Wang Hongning. Selective sampling for sensor type classification in buildings[C].//Proceedings-2020 19th ACM/IEEE International Conference on Information Processing in Sensor Networks, IPSN 2020, p.241-252.
    [16] Ng Wing W. Y., Jiang Xiaoxia, Tian Xing, Pelillo Marcello, Wang Hui, Kwong Sam Incremental hashing with sample selection using dominant sets. International Journal of Machine Learning and Cybernetics, 2020.(Article in Press)
    [17] Rahimi Hamid. Considering factors affecting the prediction of time series by improving sine-cosine algorithm for selecting the best samples in neural network multiple training model[J]. Lecture Notes in Electrical Engineering, 2019, 480:307-320.
    [18] Tang Pengfei, Du Peijun, Lin Cong, Guo Shanchuan, Qie Lu. A Novel Sample Selection Method for Impervious Surface Area Mapping Using JL1-3B Nighttime Light and Sentinel-2 Imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13:3931-3941.
    [19] Tran Nguyen, Abramenko Oleksii, Jung Alexander. On the sample complexity of graphical model selection from non-stationary samples[J]. IEEE Transactions on Signal Processing, 2020, 68:17-32.
    [20] Varshavskiy Ilyas E., Dmitriev, Ivan A., Krasnova, Anastasiia I., Polivanov Vladimir V. Selection of Sampling Rate for Digital Noise Filtering Algorithms[C].//Proceedings of the 2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering, 2020, p 932-935.
    [21] Xu Xinzheng, Li Shan, Liang Tianming, Sun Tongfeng. Sample selection-based hierarchical extreme learning machine[J]. Neurocomputing, 2020, 377:95-102.
    [22] Zhang Chenxiao, Wu Yifeng, Guo Mingming, Deng Xiaobo. Training sample selection for space-time adaptive processing based on multi-frames. Journal of Engineering[J], 2019, 20:6369-6372.
    [23] Zhang Xiwen, Seyfi Tolunay, Ju Shengtai, Ramjee Sharan, Gamal Aly El, Eldar Yonina C. Deep Learning for Interference Identification:Band, Training SNR, and Sample Selection[C]//IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC, July 2019.
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Article Metrics

Article views(296) PDF downloads(6) Cited by()

Related
Proportional views

Research on Oblique Factor Model for Selecting Training Samples

doi: 10.13203/j.whugis20200631
Funds:

the National Key Research and Development Program of China (No. 2018YFC0407804).

Abstract: Objectives: Researchers notice that the quality of training samples will impact the effective of training phase and then further will have an influence on the overall classification accuracy in the testing phase. In fact, representativeness or typicalness of training samples is able to reflect the quality of training samples in a way. Especially for the currently popular deep learning methods, it has needed thousands or millions of training samples. Therefore, how to reduce the number of training samples for deep learning method becomes a very important problem. In another hand, from the actual application angle, it is also very expensive. Therefore, we research one method of reducing the training samples as less as possible based on the representativeness or typicalness of training samples. Method: selection of training samples based on oblique factor model is proposed and it relaxes the independent condition among common factors in the orthogonal factor model, which is able to better describe the real world. Results: Experimental results show the proposed method is feasible and effective and it is able to select more representative training samples than the method of selection of training samples based on orthogonal factor model and achieve better performance in the overall classification precision and stability. Experimental results show that selection of training samples based on oblique factor model outperforms selection of training samples based on orthogonal factor model. And the distribution of selected samples becomes more decentralized and reasonable and the overall classification accuracy averagely improves about 3%. Conclusions: the proposed method, not only supports how to optimize capturing data in the theory, but also is able to guide how to effectively capture data in the actual application.

YU Xin, ZHENG Zhaobao, LI Linyi. Research on Oblique Factor Model for Selecting Training Samples[J]. Geomatics and Information Science of Wuhan University. doi: 10.13203/j.whugis20200631
Citation: YU Xin, ZHENG Zhaobao, LI Linyi. Research on Oblique Factor Model for Selecting Training Samples[J]. Geomatics and Information Science of Wuhan University. doi: 10.13203/j.whugis20200631
Reference (23)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return