Optimal Support Vector Regression Algorithms for Multifunctional Sensor Signal Reconstruction
Abstract
The empirical risk minimization methods were often used to estimate the multifunctional sensor regression function in signal reconstruction. The small size of sample data would lead to the problem of poor generalization capability and overfitting. Support vector machine (SVM) is a novel machine learning method based on structural risk minimization, and it can improve generalization capability and restrain overfitting. In this paper, an optimal ν Support Vector Regression (ν-SVR) algorithms have been proposed for multifunctional sensor reconstruction, which combined ν-SVR with particle swarm optimization (PSO), achieving accurate estimation of both the hyperparameters and reconstruction function. The results of emulation and theory analysis indicate that the proposed algorithm is more accurate and reliable for signal reconstruction.
DOI : http://dx.doi.org/10.11591/telkomnika.v12i4.4204
Keywords
Full Text:
PDFRefbacks
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Indonesian Journal of Electrical Engineering and Computer Science (IJEECS)
p-ISSN: 2502-4752, e-ISSN: 2502-4760
This journal is published by the Institute of Advanced Engineering and Science (IAES) in collaboration with Intelektual Pustaka Media Utama (IPMU).