Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data
Imbalanced learning problems are a challenge faced by classifiers when data samples have an unbalanced distribution in each class. Furthermore, the synthetic oversampling method (SMOTE) is a preprocessing technique widely used to synthesize new data and balance the different numbers of samples in ea...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2021
|
| Subjects: |
| _version_ | 1826050422287106048 |
|---|---|
| author | Pradipta, G.A. Wardoyo, R. Musdholifah, A. Sanjaya, I.N.H. |
| author_facet | Pradipta, G.A. Wardoyo, R. Musdholifah, A. Sanjaya, I.N.H. |
| author_sort | Pradipta, G.A. |
| collection | UGM |
| description | Imbalanced learning problems are a challenge faced by classifiers when data samples have an unbalanced distribution in each class. Furthermore, the synthetic oversampling method (SMOTE) is a preprocessing technique widely used to synthesize new data and balance the different numbers of samples in each class. One of the SMOTE method's expansions is based on the initial selection approach, which determines the best candidates to be oversampled in the data before the process of synthetic example generation starts. However, SMOTE and most of the existing oversampling methods based on initial selection still found overlapping data on the final result. This issue makes it difficult for any classifiers to determine the decision boundary of each class. Therefore, this research proposes a new oversampling technique called Radius-SMOTE, which emphasizes the initial selection approach by creating synthetic data based on a safe radius distance. Furthermore, new synthetic data are prevented from overlapping in the opposite class with the safe radius distance. The Radius-SMOTE was evaluated extensively with thirteen artificial imbalanced datasets from the KEEL repository. The experimental results show that the proposed method is able to achieve the best results on 5 datasets, namely yeast-1-4-5-8vs7, ecoli-0-1-3-7vs2-6, Umbilical cord, Pima, and Haberman dataset in term of various assessment metrics. Besides that, the computational cost for our proposed method is also relatively low, with an average time of 0.5 to 1 second on the 13 tested datasets. © 2013 IEEE. |
| first_indexed | 2024-03-14T00:03:43Z |
| format | Article |
| id | oai:generic.eprints.org:280330 |
| institution | Universiti Gadjah Mada |
| last_indexed | 2024-03-14T00:03:43Z |
| publishDate | 2021 |
| publisher | Institute of Electrical and Electronics Engineers Inc. |
| record_format | dspace |
| spelling | oai:generic.eprints.org:2803302023-11-10T05:52:05Z https://repository.ugm.ac.id/280330/ Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data Pradipta, G.A. Wardoyo, R. Musdholifah, A. Sanjaya, I.N.H. Structural Geology Imbalanced learning problems are a challenge faced by classifiers when data samples have an unbalanced distribution in each class. Furthermore, the synthetic oversampling method (SMOTE) is a preprocessing technique widely used to synthesize new data and balance the different numbers of samples in each class. One of the SMOTE method's expansions is based on the initial selection approach, which determines the best candidates to be oversampled in the data before the process of synthetic example generation starts. However, SMOTE and most of the existing oversampling methods based on initial selection still found overlapping data on the final result. This issue makes it difficult for any classifiers to determine the decision boundary of each class. Therefore, this research proposes a new oversampling technique called Radius-SMOTE, which emphasizes the initial selection approach by creating synthetic data based on a safe radius distance. Furthermore, new synthetic data are prevented from overlapping in the opposite class with the safe radius distance. The Radius-SMOTE was evaluated extensively with thirteen artificial imbalanced datasets from the KEEL repository. The experimental results show that the proposed method is able to achieve the best results on 5 datasets, namely yeast-1-4-5-8vs7, ecoli-0-1-3-7vs2-6, Umbilical cord, Pima, and Haberman dataset in term of various assessment metrics. Besides that, the computational cost for our proposed method is also relatively low, with an average time of 0.5 to 1 second on the 13 tested datasets. © 2013 IEEE. Institute of Electrical and Electronics Engineers Inc. 2021 Article PeerReviewed Pradipta, G.A. and Wardoyo, R. and Musdholifah, A. and Sanjaya, I.N.H. (2021) Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data. IEEE Access, 9. pp. 74763-74777. https://www.scopus.com/inward/record.uri?eid=2-s2.0-85105844830&doi=10.1109%2fACCESS.2021.3080316&partnerID=40&md5=7e77620750be71b98b44fa9206f1ed38 |
| spellingShingle | Structural Geology Pradipta, G.A. Wardoyo, R. Musdholifah, A. Sanjaya, I.N.H. Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data |
| title | Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data |
| title_full | Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data |
| title_fullStr | Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data |
| title_full_unstemmed | Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data |
| title_short | Radius-SMOTE: A New Oversampling Technique of Minority Samples Based on Radius Distance for Learning from Imbalanced Data |
| title_sort | radius smote a new oversampling technique of minority samples based on radius distance for learning from imbalanced data |
| topic | Structural Geology |
| work_keys_str_mv | AT pradiptaga radiussmoteanewoversamplingtechniqueofminoritysamplesbasedonradiusdistanceforlearningfromimbalanceddata AT wardoyor radiussmoteanewoversamplingtechniqueofminoritysamplesbasedonradiusdistanceforlearningfromimbalanceddata AT musdholifaha radiussmoteanewoversamplingtechniqueofminoritysamplesbasedonradiusdistanceforlearningfromimbalanceddata AT sanjayainh radiussmoteanewoversamplingtechniqueofminoritysamplesbasedonradiusdistanceforlearningfromimbalanceddata |