602: JEA Illustrates How to Improve Meter Estimates Using the Gradient Boosting Machine Learning Technique
Date & Time
Thursday, November 2, 2023, 10:15 AM - 11:00 AM
Greg Harvey

Gradient Boosting Machines are powerful ensemble learning algorithms that build a predictive model by sequentially adding weak learners (typically decision trees) and adjusting their weights to minimize prediction errors. Each subsequent tree focuses on correcting the errors of the previous ones, resulting in a highly accurate and robust predictive model.

JEA improves meter estimation techniques through machine learning methods they design to continuously improve on models that use averages or traditional regression to calculate these meter estimates. JEA demonstrates how to overcome computational inefficiencies with the algorithm they developed using XGBoost, an open-source gradient boosting library designed specifically to be efficient, flexible, and portable through parallel tree boosting.

Session Takeaways:

  • Data Processing
  • Model development
  • Interpreting and delivering results
Location Name
Tuscan II
Full Address
Loews Portofino Bay Hotel at Universal Orlando
5601 Universal Blvd
Orlando, FL 32819
United States
Focus Area
Data Science
Presentation 1