A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification

Bo Zhou, Cheng Yang, Haixiang Guo, Takayuki Furuzuki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

This paper focuses on imbalanced dataset classification problem by using SVM and oversampling method. Traditional oversampling method increases the occurrence of over-lapping between classes, which leads to poor generalization of SVM classification. To solve this problem this paper proposes a combined method of quasi-linear SVM and assembled SMOTE. The quasi-linear SVM is an SVM with quasi-linear kernel function. It realizes an approximate nonlinear separation boundary by mulit-local linear boundaries with interpolation. The assembled SMOTE implements oversampling with considering of the data distribution information and avoids occurrence of overlapping between classes. Firstly, a partition method based on Minimal Spanning Tree is proposed to obtain local linear partitions, each of which can be separated with one linear separation boundary. Secondly, using the information of local linear partitions, the assembled SMOTE generates synthetic minority class samples. Finally, the quasi-linear SVM realizes a classification of oversampled datasets in the same way as a standard SVM by using a composite quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.

Original languageEnglish
Title of host publicationProceedings of the International Joint Conference on Neural Networks
DOIs
Publication statusPublished - 2013
Event2013 International Joint Conference on Neural Networks, IJCNN 2013 - Dallas, TX
Duration: 2013 Aug 42013 Aug 9

Other

Other2013 International Joint Conference on Neural Networks, IJCNN 2013
CityDallas, TX
Period13/8/413/8/9

Fingerprint

Lapping
Interpolation
Composite materials
Experiments

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Zhou, B., Yang, C., Guo, H., & Furuzuki, T. (2013). A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification. In Proceedings of the International Joint Conference on Neural Networks [6707035] https://doi.org/10.1109/IJCNN.2013.6707035

A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification. / Zhou, Bo; Yang, Cheng; Guo, Haixiang; Furuzuki, Takayuki.

Proceedings of the International Joint Conference on Neural Networks. 2013. 6707035.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhou, B, Yang, C, Guo, H & Furuzuki, T 2013, A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification. in Proceedings of the International Joint Conference on Neural Networks., 6707035, 2013 International Joint Conference on Neural Networks, IJCNN 2013, Dallas, TX, 13/8/4. https://doi.org/10.1109/IJCNN.2013.6707035
Zhou B, Yang C, Guo H, Furuzuki T. A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification. In Proceedings of the International Joint Conference on Neural Networks. 2013. 6707035 https://doi.org/10.1109/IJCNN.2013.6707035
Zhou, Bo ; Yang, Cheng ; Guo, Haixiang ; Furuzuki, Takayuki. / A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification. Proceedings of the International Joint Conference on Neural Networks. 2013.
@inproceedings{267a2c173709418f943ee49327f7a6d6,
title = "A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification",
abstract = "This paper focuses on imbalanced dataset classification problem by using SVM and oversampling method. Traditional oversampling method increases the occurrence of over-lapping between classes, which leads to poor generalization of SVM classification. To solve this problem this paper proposes a combined method of quasi-linear SVM and assembled SMOTE. The quasi-linear SVM is an SVM with quasi-linear kernel function. It realizes an approximate nonlinear separation boundary by mulit-local linear boundaries with interpolation. The assembled SMOTE implements oversampling with considering of the data distribution information and avoids occurrence of overlapping between classes. Firstly, a partition method based on Minimal Spanning Tree is proposed to obtain local linear partitions, each of which can be separated with one linear separation boundary. Secondly, using the information of local linear partitions, the assembled SMOTE generates synthetic minority class samples. Finally, the quasi-linear SVM realizes a classification of oversampled datasets in the same way as a standard SVM by using a composite quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.",
author = "Bo Zhou and Cheng Yang and Haixiang Guo and Takayuki Furuzuki",
year = "2013",
doi = "10.1109/IJCNN.2013.6707035",
language = "English",
isbn = "9781467361293",
booktitle = "Proceedings of the International Joint Conference on Neural Networks",

}

TY - GEN

T1 - A quasi-linear SVM combined with assembled SMOTE for imbalanced data classification

AU - Zhou, Bo

AU - Yang, Cheng

AU - Guo, Haixiang

AU - Furuzuki, Takayuki

PY - 2013

Y1 - 2013

N2 - This paper focuses on imbalanced dataset classification problem by using SVM and oversampling method. Traditional oversampling method increases the occurrence of over-lapping between classes, which leads to poor generalization of SVM classification. To solve this problem this paper proposes a combined method of quasi-linear SVM and assembled SMOTE. The quasi-linear SVM is an SVM with quasi-linear kernel function. It realizes an approximate nonlinear separation boundary by mulit-local linear boundaries with interpolation. The assembled SMOTE implements oversampling with considering of the data distribution information and avoids occurrence of overlapping between classes. Firstly, a partition method based on Minimal Spanning Tree is proposed to obtain local linear partitions, each of which can be separated with one linear separation boundary. Secondly, using the information of local linear partitions, the assembled SMOTE generates synthetic minority class samples. Finally, the quasi-linear SVM realizes a classification of oversampled datasets in the same way as a standard SVM by using a composite quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.

AB - This paper focuses on imbalanced dataset classification problem by using SVM and oversampling method. Traditional oversampling method increases the occurrence of over-lapping between classes, which leads to poor generalization of SVM classification. To solve this problem this paper proposes a combined method of quasi-linear SVM and assembled SMOTE. The quasi-linear SVM is an SVM with quasi-linear kernel function. It realizes an approximate nonlinear separation boundary by mulit-local linear boundaries with interpolation. The assembled SMOTE implements oversampling with considering of the data distribution information and avoids occurrence of overlapping between classes. Firstly, a partition method based on Minimal Spanning Tree is proposed to obtain local linear partitions, each of which can be separated with one linear separation boundary. Secondly, using the information of local linear partitions, the assembled SMOTE generates synthetic minority class samples. Finally, the quasi-linear SVM realizes a classification of oversampled datasets in the same way as a standard SVM by using a composite quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.

UR - http://www.scopus.com/inward/record.url?scp=84893527782&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893527782&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2013.6707035

DO - 10.1109/IJCNN.2013.6707035

M3 - Conference contribution

AN - SCOPUS:84893527782

SN - 9781467361293

BT - Proceedings of the International Joint Conference on Neural Networks

ER -