An algorithmic approach that aims to reduce the number of variables (features) in a data set to be used for training a model while still keeping as much information as possible. With Principal Component Analysis, variables are transformed into a new set of variables, called principal components, which are linear combination of original variables. These principal components are arranged such that the first principal component accounts for most of the possible variation of original data, and each additional component has the highest possible variance. In this way you can see which aspects of the data have the most impact and which have the least, and remove those features or dimensions that have the least impact.
Glossary of Artificial Intelligence (AI), Machine Learning (ML), and Big Data Terms
Principal Component Analysis (PCA)
