基于FDR的证据理论改进算法

Improved algorithm of evidence theory based on FDR

ES评分 0

DOI 10.12208/j.aics.20230030
刊名
Advances in International Computer Science
年,卷(期) 2023, 3(4)
作者
作者单位

Big Data Analytics Trading Inc. 美国 ;

摘要
为简化证据理论合成规则融合过程,提高其融合效果,本文应用特征降维(Feature Dimension Reduction,FDB)技术,提出一种行之有效的证据理论改进算法。实验结果表明:基于FDR的证据理论改进算法具有融合过程简单、融合效果好、类型识别率高等特点,该算法经过数据集测试后,其类型识别率升高至94%,完全符合实际应用需求。希望通过这次研究,为相关人员提供有效的借鉴和参考。
Abstract
In order to simplify the fusion process of evidence theory synthesis rules and improve its fusion effect, this paper applies Feature Dimension Reduction (FDB) technology to propose an effective evidence theory improvement algorithm. The experimental results show that the improved algorithm based on FDR evidence theory has the characteristics of simple fusion process, good fusion effect, and high type recognition rate. After being tested on the dataset, the type recognition rate of the algorithm increased to 94%, fully meeting the practical application requirements. I hope to provide effective reference and guidance for relevant personnel through this study.
关键词
证据理论;组合规则;样本分类
KeyWord
Evidence theory; Combination rules; Sample classification
基金项目
页码 10-13
  • 参考文献
  • 相关文献
  • 引用本文

丁烈骁*. 基于FDR的证据理论改进算法 [J]. 国际计算机科学进展. 2023; 3; (4). 10 - 13.

  • 文献评论

相关学者

相关机构