The Intrinsic Dimensionality of Spectropolarimetric Data
The Intrinsic Dimensionality of Spectropolarimetric Data
复制标题
光谱偏振数据的内在维度
DOI：
10.1086/513069
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
M. J. M. González
中科院分区：
文献类型：
--
作者：
A. A. Ramos;H. Socas;A. Ariste;M. J. M. González;M. J. M. González;M. J. M. González
估计了分光偏振数据中可用的信息量。为此，数据的内在维数推断与最近派生的估计的基础上最近的邻居的考虑和获得应用最大似然原理的帮助。我们详细地表明，估计正确地捕捉人工数据集的固有尺寸与已知的尺寸。噪声在估计维度的影响进行了深入的分析，我们得出结论，它引入了一个积极的偏见，需要考虑。本文还详细研究了630 nm可见光和1.5 μm近红外波段的真实的同步偏振光谱观测，表明近红外波段的偏振光谱数据比可见光波段的偏振光谱数据提供了更多的太阳大气物理条件的信息。最后，我们证明了存在于观测数据集中的信息量是可用谱线数量的单调递增函数。
The amount of information available in spectropolarimetric data is estimated. To this end, the intrinsic dimensionality of the data is inferred with the aid of a recently derived estimator based on nearest neighbor considerations and obtained applying the principle of maximum likelihood. We show in detail that the estimator correctly captures the intrinsic dimension of artificial data sets with known dimension. The effect of noise in the estimated dimension is analyzed thoroughly, and we conclude that it introduces a positive bias that needs to be accounted for. Real simultaneous spectropolarimetric observations in the visible 630 nm and the near-infrared 1.5 μm spectral regions are also investigated in detail, showing that the near-infrared data set provides more information of the physical conditions in the solar atmosphere than the visible data set. Finally, we demonstrate that the amount of information present in an observed data set is a monotonically increasing function of the number of available spectral lines.