The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Diagnostic Visualization for Machine Learning with YellowBrick w/ Rebecca Bilbro - TWiML Talk #264
May 10, 2019
Rebecca Bilbro, Head of Data Science at ICX Media and co-creator of the YellowBrick visualization library, shares her fascinating journey in toolmaking. She discusses how YellowBrick enhances machine learning visualization and its innovative applications, including natural language processing. Listeners will learn about the significance of part of speech ratios in categorizing document types and how classification techniques can bolster regression analysis. Bilbro also emphasizes the role of visual tools in model selection and performance tracking.
AI Snips
Chapters
Transcript
Episode notes
Rebecca's Path to Data Science
- Rebecca Bilbro's background is in mathematics, English, and visualization practices in engineering.
- She transitioned into data science during the Obama administration by analyzing messy government data and building tools in Python.
Open Source as Activism
- Building tools and contributing to open source is less about public service and more about activism or art.
- It's about making the change one wants to see in the world.
Yellowbrick's Origin
- Yellowbrick was initially created to help students understand machine learning decisions through visualizations.
- The project expanded to help engineers and data scientists diagnose and visualize machine learning processes.