Рет қаралды 625
Date Presented: 12/08/2023
Speaker: Sean Taylor, Motif Analytics
Abstract:
A variety of business processes can be captured and represented as event sequences, especially as product instrumentation becomes more comprehensive in web and mobile applications. However, low-level event data are high dimensional and inherently challenging to wrangle, model, and visualize. The result is that analytics professionals typically aggregate data before visualization and estimation, discarding potentially valuable information and introducing bias. In this talk I discuss promising approaches we are applying for studying event sequences, with a focus on exploratory analysis and hypothesis generation tasks. I will draw some interesting connections to useful methodologies: causal inference techniques using panel data, deep learning architectures for dimensionality reduction, and generative AI for summarizing long and complex sequences.
Speaker's bio:
Sean J. Taylor is co-founder and chief scientist at Motif Analytics. Previously he was a data scientist and head of Lyft's Rideshare Labs and spent seven years as a research scientist on Facebook's Core Data Science team. Sean's work is at the intersection of experimentation and causal inference, with a focus on applied problems and generating business value using the latest methods. He earned his PhD in Information Systems from NYU’s Stern School of Business as well as a BS in Economics from Wharton.