Knowledge Network Node

Granular Ball Label Noise Filtering Method for RegressionChinese Full TextEnglish Full Text (MT)

SHI Ying;QI Xiaobo;QI Hui;JIANG Gao-xia;WANG Wen-jian;College of Computer Science and Technology, Taiyuan Normal University;School of Computer and Information Technology, Shanxi University;Institute of Intelligent Information Processing, Shanxi University;Department of Network Security, Shanxi Police College;

Abstract: The performance of machine learning algorithms is closely related to data quality.However, real datasets may have various types of noise.How to deal with label noise is one of the key challenges in machine learning.Due to the complexity of regression label noise, existing noise filtering methods are not effective enough in dealing with numerical label noise.A granular ball estimate of numerical label noise is developed thorough label discretization.Then the generalization error estimation of reg... More
  • DOI:

    10.20009/j.cnki.21-1106/TP.2023-0495

  • Series:

    (I) Electronic Technology & Information Science

  • Subject:

    Automation Technology

  • Classification Code:

    TP181

Download the mobile appuse the app to scan this coderead the article.

Tips: Please download CAJViewer to view CAJ format full text.

Download: 103 Page: 1832-1838 Pagecount: 7 Size: 1442K

Related Literature
  • Similar Article
  • Reader Recommendation
  • Associated Author