Scale-Invariant Geometric Data Analysis (SIGDA)
Marina Girgis, Max Robinson · DigitalCommons - CalPoly (California State Polytechnic University) · 2018
The purpose of this research is to introduce a new data analysis method called Scale Invariant Geometric Data Analysis (SIGDA). SIGDA has been shown to be more informative than more common data analysis methods, such as Principal Component Analysis (PCA). SIGDA is used to visualize complex data sets in a way that accurately preserves data patterns and behavior. SIGDA is designed to preserve relative ratios in a numerical matrix, and the number of entries has to be more than the total number of rows and columns. Our research involved providing a simple explanation of SIGDA's mathematical process—simple enough for the public to understand—and constructing educational materials to promote the use of SIGDA. I worked with my mentor, Max Robinson, to create posters and presentations to illustrate how SIGDA works. We used feedback from fellow scientists to continue to update and simplify the material to a level that a high school student could understand.