Open Access Senior Thesis
Bachelor of Science
© 2017 Kathryn A Dover
Finding patterns in high dimensional data can be difficult because it cannot be easily visualized. There are many different machine learning methods to fit data in order to predict and classify future data but there is typically a large expense on having the machine learn the fit for a certain part of a dataset. We propose a geometric way of defining different patterns in data that is invariant under size and rotation. Using a Gaussian Process, we find that pattern within stock datasets and make predictions from it.
Dover, Kathryn, "Pattern Recognition in Stock Data" (2017). HMC Senior Theses. 105.