Your browser doesn't support javascript.
loading
Distribution-Sensitive Unbalanced Data Oversampling Method for Medical Diagnosis.
Han, Weihong; Huang, Zizhong; Li, Shudong; Jia, Yan.
Afiliação
  • Han W; Institute of Advanced Technology in Cyberspace, Guangzhou University, Guangzhou, 510006, Guangdong, China. hanweihong@gzhu.edu.cn.
  • Huang Z; Institute of Electronic and Information Engineering of UESTC in Guangdong, Guangzhou, Guangdong, China. hanweihong@gzhu.edu.cn.
  • Li S; School of Computer of National University of Defense Technology, Changsha, 410073, Hunan, China.
  • Jia Y; Institute of Advanced Technology in Cyberspace, Guangzhou University, Guangzhou, 510006, Guangdong, China.
J Med Syst ; 43(2): 39, 2019 Jan 10.
Article em En | MEDLINE | ID: mdl-30631957
ABSTRACT
Aiming at the problem of low accuracy of classification learning algorithm caused by serious imbalance of sample set in medical diagnostic application, this paper proposes a distribution-sensitive oversampling algorithm for imbalanced data. The algorithm accurately divides the minority samples into noise samples, unstable samples, boundary samples and stable samples according to the location of the minority samples. Different samples are processed differently to select the most suitable sample for the synthesis of new samples. In the case of sample synthesis, a distribution-sensitive sample synthesis method is adopted. Different sample synthesis methods are selected according to their different distance from the surrounding minority samples, so as to ensure that the newly synthesized samples have the same characteristics with the original minority samples. The real medical diagnostic data test shows that this algorithm improves the accuracy rate of classification learning algorithm compared with the existing sampling algorithms, especially for the accuracy rate and recall rate of minority classes.
Assuntos
Palavras-chave

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Assunto principal: Algoritmos / Interpretação Estatística de Dados / Diagnóstico / Big Data Tipo de estudo: Diagnostic_studies / Prognostic_studies Limite: Humans Idioma: En Revista: J Med Syst Ano de publicação: 2019 Tipo de documento: Article País de afiliação: China

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Assunto principal: Algoritmos / Interpretação Estatística de Dados / Diagnóstico / Big Data Tipo de estudo: Diagnostic_studies / Prognostic_studies Limite: Humans Idioma: En Revista: J Med Syst Ano de publicação: 2019 Tipo de documento: Article País de afiliação: China