Principal component analysis
Principal component analysis: A linear method projecting data onto orthogonal directions of maximum variance to reduce dimensions and redundancy. It requires consistent scaling and treats variance as information. Principal components may lose nonlinear structure or relevance for the target. This perspective connects the definition with practical behaviour in artificial intelligence systems.
In italiano: Analisi delle Componenti PrincipaliLegge I · Embedding, feature e spazi vettorialiIt requires consistent scaling and treats variance as information. Principal components may lose nonlinear structure or relevance for the target.