Threshold Optimization Algorithm -KSVM for Unbalanced Data Classification Prediction

DOI

10.2991/msam-17.2017.25How to use a DOI?

Keywords

SVM; KNN; threshold; imbalanced datasets; Genetic Algorithm

Abstract

In the era of Big Data, the multidimensional datasets from electric power, medical treatment and other industries are often unbalanced, and the positive data usually costs more seriously when being classified. According to different datasets, the distribution tendency of datasets may reduce the accuracy of classifiers. Based on SVM, the traditional Classifier KSVM introduces KNN algorithm effectively, that increases the effective classification information for error-prone points near the hyperplane, but at the same time it introduces more noise. Based on the defect that the KSVM algorithm with fixed threshold applied to unbalanced datasets, this paper proposes an improved -KSVM classifier with thresholds of dynamic adjustment for different datasets. The classifier applies Genetic Algorithm to adjust the boundary, namely the threshold dynamically so that the misclassification information is reduced. The experimental results show that the prediction accuracy is improved greatly.

Copyright

© 2017, 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  - Guoqing Ge
AU  - Xin Jin
AU  - Xu Lu
AU  - Yongbin Zhao
PY  - 2017/03
DA  - 2017/03
TI  - Threshold Optimization Algorithm -KSVM for Unbalanced Data Classification Prediction
BT  - Proceedings of the 2017 2nd International Conference on Modelling, Simulation and Applied Mathematics (MSAM2017)
PB  - Atlantis Press
SP  - 109
EP  - 113
SN  - 1951-6851
UR  - https://doi.org/10.2991/msam-17.2017.25
DO  - 10.2991/msam-17.2017.25
ID  - Ge2017/03
ER  -