Principal component analysis (PCA) is a dimensionality reduction technique in which a high-dimensional dataset is projected onto directions of decreasing importance, ordered by the amount of variance explained. The projected data has lower rank and is thus easier to analyze. This in turn can be helpful in a variety of problems such as extracting information, data compression or analyzing structures hidden in the data.
📄 Check out my note on principal component analysis here.
💻 The Python code shown in this note can be accessed here.
This note is a tutorial on PCA with a deeper focus on the linear algebra and the statistics governing this method. My aim is to present a deep understanding of several linear algebra concepts. My goal is also to gather all important knowledge on PCA in one place.