An Improved AdaBoost-SVM Model Based on Sample Weights and Sampling Equilibrium

DOI

10.2991/iceeecs-16.2016.9How to use a DOI?

Keywords

AdaBoost,SVM, Sample Weights, Sampling Equilibrium

Abstract

The existing model which combines AdaBoost and SVM has poor performance when dealing with the imbalance dataset in multi-label classification. To deal with this problem, we proposed a new model SAB-WSVM. In our model, we modified AdaBoost original sampling methods in order to make it more balanced and more informative. Also we combined the sample weights of SVM and weights of AdaBoost to make SVM pay more attention to the samples which are difficult to be classified.In the experiment, we test it with two datasets. The results show that our model has better performance in the unbalanced multi-label datasets.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hongchen Guo
AU  - Junbang Ma
AU  - Zhiqiang Li
PY  - 2016/12
DA  - 2016/12
TI  - An Improved AdaBoost-SVM Model Based on Sample Weights and Sampling Equilibrium
BT  - Proceedings of the 2016 4th International Conference on Electrical & Electronics Engineering and Computer Science (ICEEECS 2016)
PB  - Atlantis Press
SP  - 34
EP  - 38
SN  - 2352-538X
UR  - https://doi.org/10.2991/iceeecs-16.2016.9
DO  - 10.2991/iceeecs-16.2016.9
ID  - Guo2016/12
ER  -