#noindex
많은 양의 데이터를 데이터를 본질을 해치지 않는 저차원으로 압축하는 것
Method for compressing a lot of data into something that captures the essence of the original data

차원축소 [[dimensionality_reduction]],
특징추출 [[feature_extraction]] // [[특징,feature]]을 ...
기법.
[[분산,variance]]은 최대한 보존하면서
[[축,axis]] or [[기저,basis]]을 새로 찾는...

한국어 설명:
https://angeloyeo.github.io/2019/07/27/PCA.html
https://darkpgmr.tistory.com/110

mklink
[[고유값,eigenvalue]]
[[고유벡터,eigenvector]]
[[공분산행렬,covariance_matrix]] curr see [[공분산,covariance]]
Kosambi–Karhunen–Loève theorem ... or
Karhunen–Loève theorem .... KKL_theorem KL_theorem ?
{
'''Kosambi–Karhunen–Loève theorem'''

//wpen
[[확률과정,stochastic_process]]론의 [[정리,theorem]].
확률과정을 [[직교함수,orthogonal_function]]들의 무한한 [[선형결합,linear_combination]]으로 [[표현,representation]]하는 것에 대한 -
analogous to: [[함수,function]]을 bounded_interval(유계구간? curr see [[구간,interval]]) [[푸리에_급수,Fourier_series]] ddddddddd

[[WpEn:Kosambi–Karhunen–Loève_theorem]]

Naver:"kkl 정리" Ggl:"kkl 정리" Google:kkl.theorem
Up: [[정리,theorem]]
}

= tmp links ko =
http://t-robotics.blogspot.com/2018/02/32-pca.html
https://www.samsungsds.com/kr/insights/mathematics_for_ML.html 65% 쯤. "주성분 분석에서의 응용" 검색.

= tmp links en =
Principal Component Analysis Explained Visually
https://setosa.io/ev/principal-component-analysis/
https://news.ycombinator.com/item?id=27017675

= tmp =

tmp video
{
StatQuest: Principal Component Analysis 주성분 분석 (PCA)
https://www.youtube.com/watch?v=FgakZw6K1QQ

KU김성범
https://www.youtube.com/watch?v=FhQm2Tc8Kic
}

Compare:
[[독립성분분석,independent_component_analysis,ICA]]

[[주성분,principal_component]] { Up: [[성분,component]] ? }

----
[[Zeta:주성분_분석_PCA]]
[[WpKo:주성분_분석]]
[[WpEn:Principal_component_analysis]]

Up: [[자료,data]]??
 [[분석,analysis]] > [[다변량분석,multivariate_analysis]]
 [[성분분석,component_analysis]]? ( [[성분,component]]? ) (이건 아마 안 만들어질 듯.)