A Combined Algorithm for Imbalanced Classification Based on Dual Distribution Representation Learning and Classifier Decoupling Learning
Existing classification algorithms for imbalanced datasets adopt data resampling, classes reweighting and other class balancing strategies to strengthen representation ability for minority classes and adjust the classification interface. However, these algorithms weaken the network's representa...
Saved in:
Published in | 2022 IEEE 2nd International Conference on Information Communication and Software Engineering (ICICSE) pp. 18 - 24 |
---|---|
Main Authors | , , , |
Format | Conference Proceeding |
Language | English |
Published |
IEEE
18.03.2022
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Existing classification algorithms for imbalanced datasets adopt data resampling, classes reweighting and other class balancing strategies to strengthen representation ability for minority classes and adjust the classification interface. However, these algorithms weaken the network's representation ability for majority classes. Therefore, a combined algorithm is proposed based on dual distribution representation learning (DDRL) and classifier decoupling learning (CDL). Here, DDRL preserves the original distribution and samples the balanced distribution from it to guide the learning of dual distribution representation, which enhances minority classes' feature representation ability and retains it for majority classes. CDL decouples the classifier from feature representation network, and trains an MLP classifier with a balanced subset, aiming at adjusting the classification deviation caused by weak features of minority classes. Experimental results show that the proposed algorithm can improve the classification accuracy on class imbalanced datasets effectively. |
---|---|
AbstractList | Existing classification algorithms for imbalanced datasets adopt data resampling, classes reweighting and other class balancing strategies to strengthen representation ability for minority classes and adjust the classification interface. However, these algorithms weaken the network's representation ability for majority classes. Therefore, a combined algorithm is proposed based on dual distribution representation learning (DDRL) and classifier decoupling learning (CDL). Here, DDRL preserves the original distribution and samples the balanced distribution from it to guide the learning of dual distribution representation, which enhances minority classes' feature representation ability and retains it for majority classes. CDL decouples the classifier from feature representation network, and trains an MLP classifier with a balanced subset, aiming at adjusting the classification deviation caused by weak features of minority classes. Experimental results show that the proposed algorithm can improve the classification accuracy on class imbalanced datasets effectively. |
Author | Gao, Hongxia Liao, Hongyu Ma, Jianliang Lin, Guoyuan |
Author_xml | – sequence: 1 givenname: Guoyuan surname: Lin fullname: Lin, Guoyuan email: scutguoyuanlin@163.com organization: South China University of Technology,School of Automation Science and Engineering,Guangzhou,China – sequence: 2 givenname: Hongyu surname: Liao fullname: Liao, Hongyu email: hongyuliao@foxmail.com organization: South China University of Technology,School of Automation Science and Engineering,Guangzhou,China – sequence: 3 givenname: Hongxia surname: Gao fullname: Gao, Hongxia email: hxgao@scut.edu.cn organization: South China University of Technology,School of Automation Science and Engineering,Guangzhou,China – sequence: 4 givenname: Jianliang surname: Ma fullname: Ma, Jianliang email: majl@midea.com organization: Midea Central Institute,Midea Group,Foshan,China |
BookMark | eNpFkE1OwzAQhY0ECyg9ARtfIME_TWIvQ1JopEhI_KyrcTIplhI7ctIFN-DYBFrEaubN9_SkNzfk0nmHhFDOYs6Zvq-KqnjdJomUWSyYELFWQmnJLshaZ4qnabJRYgHX5CunhR-MddjSvD_4YOePgXY-0Gow0INrFlD0ME22sw3M1jv6ANNyXJbyCD0t7TQHa46_6AXHgBO6-eSsEYKz7kDB_adgoCU2_jj2P-TPckuuOugnXJ_nirw_bt-KXVQ_P1VFXkdWMDlHvG3AJCKRymgpmeJCKiZbw7NGLyWZ7ozZCJOkEjimDBl0HDVwnim1iFauyN0p1yLifgx2gPC5P_9HfgMNpmJv |
ContentType | Conference Proceeding |
DBID | 6IE 6IL CBEJK RIE RIL |
DOI | 10.1109/ICICSE55337.2022.9828930 |
DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE/IET Electronic Library IEEE Proceedings Order Plans (POP All) 1998-Present |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: RIE name: IEEE/IET Electronic Library url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
EISBN | 9781665482202 1665482206 |
EndPage | 24 |
ExternalDocumentID | 9828930 |
Genre | orig-research |
GrantInformation_xml | – fundername: Natural Science Foundation of Guangdong Province funderid: 10.13039/501100003453 |
GroupedDBID | 6IE 6IL CBEJK RIE RIL |
ID | FETCH-LOGICAL-i203t-1dcab52538b93308123803db17c998209fbb42b563a1e60e0af1e9a11788e0ad3 |
IEDL.DBID | RIE |
IngestDate | Thu Jun 29 18:36:45 EDT 2023 |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-i203t-1dcab52538b93308123803db17c998209fbb42b563a1e60e0af1e9a11788e0ad3 |
PageCount | 7 |
ParticipantIDs | ieee_primary_9828930 |
PublicationCentury | 2000 |
PublicationDate | 2022-March-18 |
PublicationDateYYYYMMDD | 2022-03-18 |
PublicationDate_xml | – month: 03 year: 2022 text: 2022-March-18 day: 18 |
PublicationDecade | 2020 |
PublicationTitle | 2022 IEEE 2nd International Conference on Information Communication and Software Engineering (ICICSE) |
PublicationTitleAbbrev | ICICSE |
PublicationYear | 2022 |
Publisher | IEEE |
Publisher_xml | – name: IEEE |
Score | 1.8492049 |
Snippet | Existing classification algorithms for imbalanced datasets adopt data resampling, classes reweighting and other class balancing strategies to strengthen... |
SourceID | ieee |
SourceType | Publisher |
StartPage | 18 |
SubjectTerms | class imbalance Classification algorithms classifier decoupling Conferences multi-classification Representation learning Software algorithms Software engineering |
Title | A Combined Algorithm for Imbalanced Classification Based on Dual Distribution Representation Learning and Classifier Decoupling Learning |
URI | https://ieeexplore.ieee.org/document/9828930 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV07T8MwELZKJyZALeItD4wkdRzbTcbSh1qkIgRU6lb5lYKgKaqShV_Az-bspq1ADGx2YiWRT87dZ393H0LXCVHcxFIFVjAesMiwQHJGg5RrDg5Q0MzLAY3vxXDC7qZ8WkM321wYa60nn9nQNf1Zvlnq0m2VtVIHD2IA6HsJoetcrQ05h6StUXfUfepziF_agPsoDavhP3RTvNsYHKDx5oVrtshbWBYq1J-_ajH-94sOUXOXoIcftq7nCNVs3kBfHQyrG5CuNbjzPl8C7H9ZYAhK8WihHINRww0vgunoQd4i-BacmMHQ6JXyHfdcFd1KAAs_eo5slZqU46oQ6xzLfPcUu8I9wK-lS-udb4c00WTQf-4Og0poIXilJC6CyGipOIV_n3L7G-Dz44TERkVtDWiMkjRTilHFRSwjK4glMotsKqMI8DN0THyM6vkytycIM8J0OxUJVan24pQ8kQrCGC5sBrGpOEUNN4uzj3UtjVk1gWd_Xz5H-86SjvMVJReoXqxKewlBQKGuvPW_AbpmtBM |
link.rule.ids | 310,311,786,790,795,796,802,27956,55107 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1NT8IwGG4IHvSkBozf9uDRja5b93FEPsIUiFFIuJF27dAIw5Dt4i_wZ_u2DIjGg7fue2m3Pe_TPe_7IHQbEsGky4WlfI9ZniM9izOPWhFLGACgT1NjBzQY-r2x9zBhkwq62-bCKKWM-EzZumn-5ctlUuipskak6YELBH0PcJ4E62ytjTyHRI24FbdeOgwimACYH6V2ecAP5xQDHN1DNNhccq0XebeLXNjJ569qjP-9pyNU36Xo4act-Byjispq6KuJ4f0Grqskbs5nSyD-rwsMYSmOF0JrGBPYYGwwtUDIjAm-BxiTGBrtgs9xW9fRLS2w8LNRyZbJSRkuS7HOMM92Z1Er3AYGW-jE3tl2lzoadzujVs8qrRasN0rc3HJkwgWj8PUTeoYDUN8NiSuFEyTAxyiJUiE8Kpjvckf5RBGeOirijgMMGhake4Kq2TJTpwh7xEuCyA-piBJjT8lCLiCQYb5KITr1z1BN9-L0Y11NY1p24Pnfq2_Qfm806E_78fDxAh3oUdUKMCe8RNV8VagrCAlycW2ehG9f97dn |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2022+IEEE+2nd+International+Conference+on+Information+Communication+and+Software+Engineering+%28ICICSE%29&rft.atitle=A+Combined+Algorithm+for+Imbalanced+Classification+Based+on+Dual+Distribution+Representation+Learning+and+Classifier+Decoupling+Learning&rft.au=Lin%2C+Guoyuan&rft.au=Liao%2C+Hongyu&rft.au=Gao%2C+Hongxia&rft.au=Ma%2C+Jianliang&rft.date=2022-03-18&rft.pub=IEEE&rft.spage=18&rft.epage=24&rft_id=info:doi/10.1109%2FICICSE55337.2022.9828930&rft.externalDocID=9828930 |