首页> 外文期刊>International Journal of Innovative Computing Information and Control >LEAST SQUARES TWIN PROJECTION SUPPORT VECTOR REGRESSION
【24h】

LEAST SQUARES TWIN PROJECTION SUPPORT VECTOR REGRESSION

机译:最小二乘投影支持向量回归

获取原文
获取原文并翻译 | 示例
           

摘要

In order to further promote the predictive performance of least squares twin support vector regression (LSTSVR), we proposed a least squares twin projection support vector regression (LSTPSVR). The variance of the projected data is integrated into the objective function of the primal problems as a new term. Minimizing the variance ensures that we can obtain a suitable projection axis, and the empirical covariance and correlation coefficients of the input and output data are embedded into the solution of linear equations automatically. Extensive experimental results on several benchmark datasets and artificial test function show that LSTPSVR can obtain better predictive performance than other state-of-the-art algorithms. In addition, the proposed algorithm is stable.
机译:为了进一步提高最小二乘双支持向量回归(LSTSVR)的预测性能,我们提出了最小二乘双投影支持向量回归(LSTPSVR)。投影数据的方差作为新项被集成到原始问题的目标函数中。使方差最小化可确保我们可以获得合适的投影轴,并且将输入和输出数据的经验协方差和相关系数自动嵌入线性方程组的解中。在多个基准数据集和人工测试功能上的大量实验结果表明,LSTPSVR可以比其他最新算法获得更好的预测性能。另外,所提出的算法是稳定的。

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