Sophie Lellis-Petrie is a Statistician at Southern California Edison. At SCE she has created experimental designs with stratified random sampling, developed predictive equipment failure models using machine learning algorithms, and conducted goal setting feasibility studies through deep dives. She combines descriptive and inferential statistics in order to have a holistic understanding of the data allowing her to fit the best models to the data. She has a Bachelor of Science in Statistics from the University of California, Los Angeles where she gained a strong foundation in statistics and was able to apply her knowledge in capstone courses in multiple disciplines. She continues to grow her statistics portfolio in the maintenance, performance and reliability group within SCE.
April 17, 2018
9:30 - 10:30
So you’ve made a model. Now what? Data science in isolation is purely academic. How do we operationalize these cutting edge machine learning models in our utilities? This talk explores what it takes to implement data science into business strategy. We discuss the process of building a production ready model, highlighting the challenges of feature … Continued