Data Science Best Practices: Prioritize Data, Think Critically, and Continuously Learn
- Prioritize data above all else - don't get distracted by flashy models or tuning
- Think critically about whether machine learning is needed to solve a problem
- Detecting outliers is a key data science task, for removal or further inspection
- Data science requires learning new tools as well as new ways of thinking
- Real-world problems encountered daily - forecasting, classification, etc.