Presented By O’Reilly and Cloudera
Make Data Work
March 5–6, 2018: Training
March 6–8, 2018: Tutorials & Conference
San Jose, CA

Want to build a better chatbot? Start with your data.

Andrew Mattarella-Micke (Intuit)
11:50am–12:30pm Wednesday, March 7, 2018
Data science and machine learning, Streaming systems and real-time applications
 Location: LL20 D
Average rating: *****
(5.00, 1 rating)

Who is this presentation for?

  • Designers, data engineers, developers, and database professionals

Prerequisite knowledge

  • Familiarity with natural language processing

What you'll learn

  • Learn how to prep data for natural language processing

Description

So you want to build a chatbot? One that is humanized, has contextual responses, and can simulate true empathy for the end users. Where do you start? Andrew Mattarella-Micke is a firm believer that a great chatbot first starts with your data. Andrew shares how Intuit’s data science team preps, cleans, organizes, and augments training data along with best practices he’s learned along the way. Andrew also discusses the unexpected sources of data and how to use data to train your chatbots for context, how to review your data to identify the blind spots, and when not to use a chatbot for your user. You’ll discover the importance of other key considerations such as cultural and language differences across demographics as well.

Photo of Andrew Mattarella-Micke

Andrew Mattarella-Micke

Intuit

Andrew Mattarella-Micke is a senior data scientist at Intuit, specializing in deep learning for NLP applications. Previously, Andrew was a postdoctoral fellow at Vanderbilt and Stanford, where he studied the brain networks underlying mathematical development. He holds a PhD in cognitive neuroscience from the University of Chicago.