Proceedings of the 2018 International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2018)

[WITHDRAWN] Boosting Relevance Vector Machine Learning Algorithm Based on Noise Self-Detection

Authors
Wangchen Qin, Fang Liu, Quan Qi, Mi Tong
Corresponding Author
Wangchen Qin
Available Online March 2018.
DOI
https://doi.org/10.2991/mecae-18.2018.29How to use a DOI?
Keywords
Relevant vector machines; AdaBoost; Noise detection.
Abstract
AdaBoost is an ensemble method to construct a strong classifier with linear combination of base classifiers, which has been applied to relevance vector machine (RVM) for performance improvement. However, the combination of the RVM and AdaBoost can be overfitting in dealing with the noisy data sets because of the inherent noise sensitivity of AdaBoost. Therefore, a boosting relevance vector machine learning algorithm based on noise self-detection is proposed, which can detect the sample type on top of the posterior probability output of RVM, remove the noisy samples and give more emphasis on the boundary samples to generate the classifiers. The algorithm was applied to real data sets, and experimental results show that the proposed method offers good performance on accuracy and generalization.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Proceedings
2018 International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2018)
Part of series
Advances in Engineering Research
Publication Date
March 2018
ISBN
978-94-6252-493-4
DOI
https://doi.org/10.2991/mecae-18.2018.29How to use a DOI?
Open Access
This is an open access article distributed under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Wangchen Qin
AU  - Fang Liu
AU  - Quan Qi
AU  - Mi Tong
PY  - 2018/03
DA  - 2018/03
TI  - [WITHDRAWN] Boosting Relevance Vector Machine Learning Algorithm Based on Noise Self-Detection
BT  - 2018 International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2018)
PB  - Atlantis Press
UR  - https://doi.org/10.2991/mecae-18.2018.29
DO  - https://doi.org/10.2991/mecae-18.2018.29
ID  - Qin2018/03
ER  -