• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>World multi-conference on systemics, cybernetics and informatics >Comparative Studies of Dimensionality Reduction Techniques for Preprocessed Correlation Based Dimensionality Reduction
【24h】

Comparative Studies of Dimensionality Reduction Techniques for Preprocessed Correlation Based Dimensionality Reduction

机译：基于预处理的降维技术降维技术的比较研究

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A new paradigm of dimensionality reduction technique is introduced. The proposed paradigm is applying the correlation to extract the relatively more correlated attributes among the original attributes as a preprocess procedure in order to interpret or manipulate the given data without losing any significant meaning for the data implementation. To compare the proposed paradigm with other dimensionality reduction techniques such as factor analysis, principal component analysis, subtractive clustering (SUBCLUST) analysis, and the various preprocessed multivariate analysis, the data of the prediction of Equity Risk Premiums (ERP) is evaluated by the reduced dimensions using the proposed paradigm applying the neurofuzzy systems for the evaluation of the predicted Equity Risk Premiums (ERP). In addition, using the results through the dimensionality reduction technique with the proposed paradigm, the statistical analyses are compared and the extracted reduced dimensions are examined.
机译：提出了一种新的降维技术范式。所提出的范例正在应用相关性以在原始属性中提取相对更相关的属性作为预处理过程，以便解释或操作给定的数据而不会丢失任何对数据实现的重要意义。为了将拟议的范式与其他降维技术（例如因子分析，主成分分析，减法聚类（SUBCLUST）分析和各种预处理的多元分析）进行比较，通过对降低的评估来评估股权风险溢价（ERP）的预测数据使用拟议的范式应用神经模糊系统对预期的股权风险溢价（ERP）进行评估。此外，使用通过降维技术的结果与所提出的范例进行比较，对统计分析进行比较并检查提取的降维。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号