Principal Component Analysis: A Primer
Principal Component Analysis is a linear algebra technique used to identify trends within a dataset and reduce its dimensionality. Statistics such as mean and covariance are referenced alongside matrix multiplication formulas so the reader may understand the work a computer must perform to analyze a dataset. Spectral decomposition and singular value decomposition of a matrix will be considered as the primary methods for an analysis. A data compression example will be given and applications in image processing will be discussed. To give an example of data projection for visualization purposes, the principal component analysis of a biological dataset will be demonstrated.
John Pike (Thesis Director)
Copyright and Permissions
Original document was submitted as an Honors Program requirement. Copyright is held by the author.
Sullivan, Simon. (2017). Principal Component Analysis: A Primer. In BSU Honors Program Theses and Projects. Item 237. Available at: https://vc.bridgew.edu/honors_proj/237
Copyright © 2017 Simon Sullivan