Driving Business Strategy with Machine Learning -- Neha Gupta Uber

Friday, October 13, 2017 - 12:30
University of San Francisco Seminar Series in Analytics
San Francisco

Driving Business Strategy with Machine Learning

In this talk, we will discuss how the Finance Data Science team at Uber uses machine learning to make strategic business decisions. While some companies tackle business strategy primarily by analyzing Excel outcomes, Uber uses finance engineers and data scientists to implement scalable frameworks to run not just yearly/half-yearly business planning, but also solve important financial problems such as call center and spend optimization using state of the art algorithmic techniques. While each element of this problem demands a unique solution, the underlying goal across our frameworks is to facilitate a more seamless planning experience and improve ROI. As data scientists, we consider the finance data science team to be guardians of Uber's P&L by giving accurate and unbiased estimates that fuel strategic decisions.
In this presentation, we will focus on the machine learning techniques and frameworks used to solve the following problems: 1) automated long-term and short-term forecasting for financial metrics and call centers, 2) customer lifetime value  3) user-level forecasting using deep learning

Bio - Neha Gupta Neha Gupta is a Data Science Manager at Uber Finance where she works on using machine learning to drive important strategic decisions for business.  Her team mainly focuses on long term as well as short term forecasting of all important financial metrics of the company. The main goal of the team is to produce the most accurate forecasts so that important decisions such as where to invest, how to spend the money, etc. can be taken in a scientific way. Before moving to Uber, Neha worked as a Sr. Data Scientist at Adobe Digital Marketing Cloud where she led research and implementation of bid optimization and personalization algorithms for advertising. She graduated with a Ph.D degree in Computer Science from University of Maryland, College Park in May, 2012. Her Ph.d thesis was on "Learning Techniques in Multi-Armed Bandits" that has several applications to A/B testing, advertising, etc. She has published several papers in the area of machine learning in various conferences and journals. She is a TED Fellow, and am deeply humbled to be part of the TED community. She finished her undergraduate studies in Electronics Engineering from Indian Institute of Technology (IIT), Varanasi, India. 


101 Howard St