Quantitative analysis of 16-17th century archaeological glass vessels using PLS regression of EPXMA and?�-XRF data

Statistical Machine Learning in Biomedical Engineering

“…Similarly, clusters 2 (rows 53 to 64 and 74 to 76) and 3 (rows 22 to 30) are depicted as blue squares and green diamonds, respectively. The detection of more than one cluster is consistent with previous studies of the dataset [48], [113] demonstrating both the successful detection of outlying sets and the prevention of their influence on the fitted PCA model by RadarTSR.…”

Statistical Machine Learning in Biomedical Engineering

“…Similarly, clusters 2 (rows 53 to 64 and 74 to 76) and 3 (rows 22 to 30) are depicted as blue squares and green diamonds, respectively. The detection of more than one cluster is consistent with previous studies of the dataset [48], [113] demonstrating both the successful detection of outlying sets and the prevention of their influence on the fitted PCA model by RadarTSR.…”

High-Dimensional Tests for Elliptical Models via Radial--Directional Dependence

“…For cookie, the blocks are high-wavelength tail windows of the biscuit-dough NIR spectra (Brown et al, 2001;Osborne et al, 1984). For glass, the blocks are local EPXMA spectral channels from the archaeological glass data used in robust sparse PCA studies (Hubert et al, 2016;Lemberge et al, 2000). The table emphasizes blocks where the radial-directional tests provide additional local or sparse-versus-dense interpretation.…”

Processing of three-dimensional microscopic X-ray fluorescence data

“…The Sr-rich volume shown in Fig. 9 consists of super-cluster volumes 3, 6, 7 and 9 and is equivalent to clusters 2, 5,8,9,10,11,12,15,16,19,24,32,34 and 36 (14 clusters in total) from the 40-means clustering. The Zr volume element (clusters 3,13,17,21,29,38) and that for the Th-rich volume (clusters 37 and 40) can be constructed in a similar way.…”

Statistical Machine Learning in Biomedical Engineering

“…Similarly, clusters 2 (rows 53 to 64 and 74 to 76) and 3 (rows 22 to 30) are depicted as blue squares and green diamonds, respectively. The detection of more than one cluster is consistent with previous studies of the dataset [48], [113] demonstrating both the successful detection of outlying sets and the prevention of their influence on the fitted PCA model by RadarTSR.…”

High-Dimensional Tests for Elliptical Models via Radial--Directional Dependence

“…For cookie, the blocks are high-wavelength tail windows of the biscuit-dough NIR spectra (Brown et al, 2001;Osborne et al, 1984). For glass, the blocks are local EPXMA spectral channels from the archaeological glass data used in robust sparse PCA studies (Hubert et al, 2016;Lemberge et al, 2000). The table emphasizes blocks where the radial-directional tests provide additional local or sparse-versus-dense interpretation.…”

Processing of three-dimensional microscopic X-ray fluorescence data

“…The Sr-rich volume shown in Fig. 9 consists of super-cluster volumes 3, 6, 7 and 9 and is equivalent to clusters 2, 5,8,9,10,11,12,15,16,19,24,32,34 and 36 (14 clusters in total) from the 40-means clustering. The Zr volume element (clusters 3,13,17,21,29,38) and that for the Th-rich volume (clusters 37 and 40) can be constructed in a similar way.…”

Statistical Machine Learning in Biomedical Engineering

“…Similarly, clusters 2 (rows 53 to 64 and 74 to 76) and 3 (rows 22 to 30) are depicted as blue squares and green diamonds, respectively. The detection of more than one cluster is consistent with previous studies of the dataset [48], [113] demonstrating both the successful detection of outlying sets and the prevention of their influence on the fitted PCA model by RadarTSR.…”

High-Dimensional Tests for Elliptical Models via Radial--Directional Dependence

“…For cookie, the blocks are high-wavelength tail windows of the biscuit-dough NIR spectra (Brown et al, 2001;Osborne et al, 1984). For glass, the blocks are local EPXMA spectral channels from the archaeological glass data used in robust sparse PCA studies (Hubert et al, 2016;Lemberge et al, 2000). The table emphasizes blocks where the radial-directional tests provide additional local or sparse-versus-dense interpretation.…”

Processing of three-dimensional microscopic X-ray fluorescence data

“…The Sr-rich volume shown in Fig. 9 consists of super-cluster volumes 3, 6, 7 and 9 and is equivalent to clusters 2, 5,8,9,10,11,12,15,16,19,24,32,34 and 36 (14 clusters in total) from the 40-means clustering. The Zr volume element (clusters 3,13,17,21,29,38) and that for the Th-rich volume (clusters 37 and 40) can be constructed in a similar way.…”