Fast, scalable and geo-distributed PCA for big data analytics

Adnan, Tamim; Tanjim, Md. Mehrab; Adnan, Muhammad Abdullah

doi:10.1016/j.is.2020.101710

Cited by 6 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PCA is an unsupervised statistical method for reducing dimensions of a database by linear combinations of a starting set of variables based on their maximum variance [37] and can convert original variables into new independent variables named principal components (PCs) [7]. It can make a preliminary judgement on the distribution status, natural aggregation, and abnormal samples [38].…”

Section: Discussionmentioning

confidence: 99%

1H NMR Combined with Multivariate Statistics for Discrimination of Female and Male Flower Buds of Populus tomentosa

et al. 2021

Molecules

View full text Add to dashboard Cite

1H Nuclear Magnetic Resonance (1H NMR) combined with multivariate statistics was adopted to discriminate female and male flower buds of Populus tomentosa in the study. Samples of 11 female and 16 male flower buds of P. tomentosa were collected in Beijing, China. 1H NMR spectra were acquired on a 400 MHz spectrometer. In total, 30 chemical compounds were identified with standards and literature according to chemical shifts, peak areas, and multiplicity. Principal component analysis (PCA), hierarchical clustering analysis (HCA), and supervised orthogonal partial least squares-discriminant analysis (OPLS-DA) were applied to discriminate female and male flower buds. An apparent grouping trend (R2X, 0.809; Q2, 0.903) between female and male groups was exhibited with PCA and HCA. The two groups were also well discriminated with OPLS-DA (R2X, 0.808; R2Y, 0.976; Q2, 0.960). Combined with variable importance in projection (VIP) > 1.0 and p < 0.05 of OPLS-DA, it was found that the content of daucosterol, β-sitosterol, ursolic acid, and betulonic acid in male group was higher than that in female, which should be the key differences of chemical constituents in female and male flower buds of P. tomentosa. The study demonstrated that 1H NMR combined with multivariate statistics could be used to discriminate female and male plants and clarify differences, which provided a novel method to identify the gender of dioecious plants.

show abstract

Section: Discussionmentioning

confidence: 99%

1H NMR Combined with Multivariate Statistics for Discrimination of Female and Male Flower Buds of Populus tomentosa

et al. 2021

Molecules

View full text Add to dashboard Cite

show abstract

“…If the dimensionality of X is large, then the computation of the eigenvalues will be time consuming, and may cause memory overflow (e.g. Adnan et al, 2021).…”

Section: Methodsmentioning

confidence: 99%

“…This limitation triggered the develop-ment of PCA extensions, including sparsity assumptions (e.g. Fan et al, 2018;Adnan et al, 2021), and massive parallelism (Lazcano et al, 2017). Battulga et al (2020) suggested a hash-tree PCA to accelerate conventional PCA by sampling similar objects while preserving the original data distribution.…”

Section: Introductionmentioning

confidence: 99%

qrpca: A Package for Fast Principal Component Analysis with GPU Acceleration

de Souza,

Quanfeng,

Shen

et al. 2022

Preprint

View full text Add to dashboard Cite

We present qrpca, a fast and scalable QR-based principal component analysis package. The software, written in both R and python languages, makes use of torch for internal matrix computations, and enables GPU acceleration, when available. qrpca provides similar functionalities to prcomp (R) and sklearn (python) packages respectively. A benchmark test shows that qrpca can achieve computational speeds 10-20 × faster for large dimensional matrices than default implementations, and is at least twice as fast for a standard decomposition of spectral data cubes. The qrpca source code is made freely available to the community.

show abstract