Knowledge Network Node

Multi-label Crowdsourced Label Inference Method Based on Label Dependence CorrectionChinese Full TextEnglish Full Text (MT)

LIU Xinyu;ZHANG Lin;JIANG Gaoxia;WANG Wenjian;School of Computer and Information Technology, Shanxi University;Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, Shanxi University;

Abstract: The rapid development of crowdsourcing platforms makes it easier to obtain data labels.However, the label quality varies a lot due to significant differences in the knowledge level and evaluation criteria of annotators.It is difficult for the annotator to provide complete and accurate data labels for each sample when obtaining labels of multi-label data through crowdsourcing, because each sample may have multiple different labels, and the number of labels may be different.To address the multi-la... More
  • DOI:

    10.20009/j.cnki.21-1106/TP.2022-0679

  • Series:

    (I) Electronic Technology & Information Science

  • Subject:

    Automation Technology

  • Classification Code:

    TP18

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

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

Download: 18 Page: 1025-1031 Pagecount: 7 Size: 1564K

Related Literature
  • Similar Article
  • Reader Recommendation
  • Associated Author