From the SciPy2013 conference, here is a compelling talk “Data Agnosticism: Feature Engineering Without Domain Expertise” by Nicholas Kridler of Accretive Health in Chicago. Nicholas is an applied mathematician, data scientist, Kaggle master, and former defense scientist specializing in data analysis and algorithmic development. In the talk, he stresses how responsible data analysis and quick iteration produce high-performing predictive models – with no domain expertise required since algorithms don’t know who generated the data. To learn more about this unique perspective, please view the presentation below:
Earn your master’s in predictive analytics completely online from Northwestern University.
Speak Your Mind