This was my Senior Thesis project at the University of Rochester.
Abstract: With future astronomical surveys expecting to see millions of transient events per night (such as LSST), there is a need to develop efficient means of identifying interesting objects. One such class of interesting objects are type Ia supernovae. I investigate the optical spectral footprint of type Ia supernovae with the goal of understanding how these objects may be identified within the spectra of their host galaxies. I describe and compare several machine learning and data-driven approaches for identifying type Ia supernovae spectroscopically. Finally, I detail the application of these methods to the Dark Energy Spectroscopic Instrument (DESI), which will observe 30 million galaxy spectra over the next ten years and as such is a prime instrument for finding supernovae and other interesting phenomena spectroscopically.