Principal component analysis constrained by layered simple structures

Yamashita, Naoto

doi:10.1007/s11634-022-00503-9

Cited by 1 publication

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It could be very challenging to show your dataset in a multidimensional hyperspace if it has more than three variables. The crucial data in a multivariate data table are extracted and visualized using principal component analysis [27]. This data is combined via PCA into a small number of new variables called principal components.…”

Section: Statistical Processing Of Datamentioning

confidence: 99%

Application of Multidimensional Statistical Methods to the Hydrochemical Study with R Software

Chaal¹,

Hamdane²,

Aboutafail³

2022

MMEP

View full text Add to dashboard Cite

Conducting extensive, time-consuming analysis campaigns is a typical technique to better understand and manage surface water quality. These usually generate a substantial amount of data that is challenging to comprehend. Principal component analysis may be advantageous for such a project (PCA). From the perspective of such an application, eight physico-chemical parameters are important: Sodium (Na+), Bicarbonate (HCO3-), Magnesium (Mg2+), Total Alkalinity (as CaCO3), Chlorides (Cl-), Potassium (K+), Calcium (Ca2+), Sulfates (SO42-), coming from the analysis of 100 water samples collected between February 2014 and December 2015 on 25 stations distributed on Inaouen catchment areas, were analyzed. The principal component analysis applied to the data showed that the variables could be grouped into two principal components. The interpretation of the results using these tools allowed us to understand that the parameters responsible for water quality are related to component Dim1 (HCO3-, CaCO3, K+, Cl-, Na+ and SO42-) and component Dim2 to the processes associated to (Ca2+ and Mg2+) for the physicochemical parameters, the Dim1 factorial design accounts for 67.80% of the variance; it is expressed towards its positive pole by HCO3-, CaCO3, K+, Cl-, Na+ and SO42-, which present good correlations between them. However, the Dim2 factorial plane represents only 17.60%, defined by the Ca2+ and Mg2+ ions towards its positive pole. The Dim1XDim2 plane's typological structure reveals the individualization of three different groupings based on their hydrochemical quality. A feasible reduction in the number of dimensions without a major loss of information was discovered by the PCA. This tool is a good choice from the standpoint of developing management tools.

show abstract

Section: Statistical Processing Of Datamentioning

confidence: 99%