Regularized Discrete Optimal Transport for Class-Imbalanced Classifications
Imbalanced class data are commonly observed in pattern analysis, machine learning, and various real-world applications. Conventional approaches often resort to resampling techniques in order to address the imbalance, which inevitably alter the original data distribution. This paper proposes a novel...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/4/524 |
| _version_ | 1797297591889690624 |
|---|---|
| author | Jiqiang Chen Jie Wan Litao Ma |
| author_facet | Jiqiang Chen Jie Wan Litao Ma |
| author_sort | Jiqiang Chen |
| collection | DOAJ |
| description | Imbalanced class data are commonly observed in pattern analysis, machine learning, and various real-world applications. Conventional approaches often resort to resampling techniques in order to address the imbalance, which inevitably alter the original data distribution. This paper proposes a novel classification method that leverages optimal transport for handling imbalanced data. Specifically, we establish a transport plan between training and testing data without modifying the original data distribution, drawing upon the principles of optimal transport theory. Additionally, we introduce a non-convex interclass regularization term to establish connections between testing samples and training samples with the same class labels. This regularization term forms the basis of a regularized discrete optimal transport model, which is employed to address imbalanced classification scenarios. Subsequently, in line with the concept of maximum minimization, a maximum minimization algorithm is introduced for regularized discrete optimal transport. Subsequent experiments on 17 Keel datasets with varying levels of imbalance demonstrate the superior performance of the proposed approach compared to 11 other widely used techniques for class-imbalanced classification. Additionally, the application of the proposed approach to water quality evaluation confirms its effectiveness. |
| first_indexed | 2024-03-07T22:22:43Z |
| format | Article |
| id | doaj.art-4b7ffd242c18404eb4d8df3ffffa07f4 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-07T22:22:43Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-4b7ffd242c18404eb4d8df3ffffa07f42024-02-23T15:26:03ZengMDPI AGMathematics2227-73902024-02-0112452410.3390/math12040524Regularized Discrete Optimal Transport for Class-Imbalanced ClassificationsJiqiang Chen0Jie Wan1Litao Ma2School of Mathematics and Physics, Hebei University of Engineering, Handan 056038, ChinaLaboratory for Space Environment and Physical Sciences, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Mathematics and Physics, Hebei University of Engineering, Handan 056038, ChinaImbalanced class data are commonly observed in pattern analysis, machine learning, and various real-world applications. Conventional approaches often resort to resampling techniques in order to address the imbalance, which inevitably alter the original data distribution. This paper proposes a novel classification method that leverages optimal transport for handling imbalanced data. Specifically, we establish a transport plan between training and testing data without modifying the original data distribution, drawing upon the principles of optimal transport theory. Additionally, we introduce a non-convex interclass regularization term to establish connections between testing samples and training samples with the same class labels. This regularization term forms the basis of a regularized discrete optimal transport model, which is employed to address imbalanced classification scenarios. Subsequently, in line with the concept of maximum minimization, a maximum minimization algorithm is introduced for regularized discrete optimal transport. Subsequent experiments on 17 Keel datasets with varying levels of imbalance demonstrate the superior performance of the proposed approach compared to 11 other widely used techniques for class-imbalanced classification. Additionally, the application of the proposed approach to water quality evaluation confirms its effectiveness.https://www.mdpi.com/2227-7390/12/4/524imbalanced dataclassificationoptimal transportmajorization–minimizationregularization term |
| spellingShingle | Jiqiang Chen Jie Wan Litao Ma Regularized Discrete Optimal Transport for Class-Imbalanced Classifications Mathematics imbalanced data classification optimal transport majorization–minimization regularization term |
| title | Regularized Discrete Optimal Transport for Class-Imbalanced Classifications |
| title_full | Regularized Discrete Optimal Transport for Class-Imbalanced Classifications |
| title_fullStr | Regularized Discrete Optimal Transport for Class-Imbalanced Classifications |
| title_full_unstemmed | Regularized Discrete Optimal Transport for Class-Imbalanced Classifications |
| title_short | Regularized Discrete Optimal Transport for Class-Imbalanced Classifications |
| title_sort | regularized discrete optimal transport for class imbalanced classifications |
| topic | imbalanced data classification optimal transport majorization–minimization regularization term |
| url | https://www.mdpi.com/2227-7390/12/4/524 |
| work_keys_str_mv | AT jiqiangchen regularizeddiscreteoptimaltransportforclassimbalancedclassifications AT jiewan regularizeddiscreteoptimaltransportforclassimbalancedclassifications AT litaoma regularizeddiscreteoptimaltransportforclassimbalancedclassifications |