Abstract
Classification is an important part of image recognition, so the accuracy and efficiency of the classifier are required. Class noise is an important factor to make the original data inseparable and seriously affects the validation accuracy of a classifier. A complete random forest method-based class noise filtering learning (CRF-NFL) is proposed, which can effectively filter out the class noise and combine with various classifiers to train the filtered training set. However, this method has two weaknesses: (1) the CRF is not optimized and cannot make the verification accuracy of the classifier better; (2) it is focused only on the combination of various classifiers. To solve these problems, an effective and robust support vector machine based on the complete random forest (CRF-ERSVM) is proposed. Firstly, the proposed method improved the complete random forest (CRF) by means of optimizing the voting percentage. Secondly, the proposed method modified the model of the SVM. At last, the effectiveness of the proposed method is verified by experiments on UCI datasets.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Martinez, C.M., Heucke, M., Wang, F.Y., Gao, B., Cao, D.: Driving style recognition for intelligent vehicle control and advanced driver assistance: a survey. IEEE Trans. Intell. Transp. Syst. 19(3), 666–676 (2017)
Xia, S., Zhang, Z., Li, W., Wang, G., Giem, E., Chen, Z.: GBNRS: a novel rough set algorithm for fast adaptive attribute reduction in classification. IEEE Trans. Knowl. Data Eng. 99, 1–1 (2020)
Xia, S., Liu, Y., Ding, X., Wang, G., Yu, H., Luo, Y.: Granular ball computing classifiers for efficient, scalable and robust learning. Inf. Sci. 483, 136–152 (2019)
Xia, S., Wu, A.: Class Noise Versus Attribute Noise.pdf, vol. 22, pp. 177–210. Kluwer Academic Publishers, Norwell, MA, USA (2014).
Garcia, L.P., Lehmann, J., de Carvalho, A.C., Lorena, A.C.: New label noise injection methods for the evaluation of noise filters. Knowl.-Based Syst., 163, 693–704 (2019).
Xia, S.Y., Xiong, Z.Y., Luo, Y.G., Dong, L.M., Xing, C.Y.: Relative density based support vector machine. Neurocomputing 149, 1424–1432 (2015)
Sluban, B., Lavrač, N.: Relating ensemble diversity and performance: a study in class noise detection. Neurocomputing 160, 120–131 (2015)
Xia, S., Wang, G., Chen, Z., Duan, Y.: Complete random forest based class noise filtering learning for improving the generalizability of classifiers. IEEE Comput. Archit. Lett. 31(11), 2063–2078 (2019)
Ho, T.K.: Proceedings of the 3rd International Conference on Document Analysis and Recognition, Montreal, QC, 14–16 August. pp. 278–282 (1995).
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
He, Q., Xia, S., Peng, X., Xia, H., Hao, Z. (2021). CRF-ERSVM: An Effective and Robust Support Vector Machine Based on Complete Random Forest. In: Jain, L.C., Kountchev, R., Shi, J. (eds) 3D Imaging Technologies—Multi-dimensional Signal Processing and Deep Learning. Smart Innovation, Systems and Technologies, vol 234. Springer, Singapore. https://doi.org/10.1007/978-981-16-3391-1_6
Download citation
DOI: https://doi.org/10.1007/978-981-16-3391-1_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-3390-4
Online ISBN: 978-981-16-3391-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)