It’s a great thing when someone hands you a well-defined machine learning problem: nice clean data, a scoring metric, and a representative test set. But the reality is often quite different. Data science teams must decide where to focus and how to apply machine learning in the best way. And when it’s time to report findings, it takes strong communication skills to be heard and get a decision.
In this webinar, John will talk about how he applied these considerations to win two analytics challenges on Kaggle sponsored by the NFL: NFL 1st and Future - Analytics and NFL Punt Analytics Competition. Analytics challenges supply data and ask participants to provide recommendations and findings. Unlike, a typical Kaggle machine learning competition, there is no objective metric or score. Reports are evaluated by a panel of judges on how well they address the issue.
After this webinar you will leave with:
- Methods to identify and prioritize opportunities for analysis
- How to apply machine learning in the context of an analytics problem
- Tips on communicating with a business audience
- Techniques to optimize the readability of Jupyter notebooks