Since the detection of an exoplanet around a sun-like star in 1995, the radial velocity (RV) method has seen tremendous improvements. Currently, we are limited in RV measurement precision due to noise from stellar variability. Detecting earth-mass exoplanets in long-period orbits around sun-like stars requires mitigating stellar activity signals caused by spots and plages. Here we test whether we can detect and remove the effects of stellar activity using machine learning (ML) techniques such as linear regression and neural networks. Previous efforts to resolve this problem have been focused on carefully filtering out activity signals in time based on frequency using a Gaussian process (Haywood et al. 2014). Instead, we separate activity signals from true Keplerian radial velocities using the shape of the cross correlation function: the summed line profile for all spectral lines. We have tested this using several ML methods on both simulated data (randomly generated with the SOAP 2.0 software (Dumusque et al. 2014)) and observations of the Sun from the HARPS-N Solar Telescope. We find that these techniques can successfully predict and remove stellar activity from radial velocity observations. Eventually, we will use these or more sophisticated methods to remove activity signals from observations of stars outside our solar system. These advancements are necessary to detect habitable-zone earth-mass exoplanets around Sun-like stars. Acknowledgement: We acknowledge the generous support from the UT Office of Undergraduate Research and the Junior Fellows Honors Program.