Case Study: Practical Label Encoding with Rainbow Method | by Anna Arakelyan | Feb, 2023
A real-world test on MassMutual’s production modelCo-authored with Dmytro KarabashPhoto by Jason Pogacnik on UnsplashIn our previous article, “Hidden Data Science Gem: Rainbow Method for Label Encoding”, we discussed the advantages of using label encoding over one-hot encoding for categorical variables, especially when developing tree-based models. We introduced the Rainbow method, which helps identify the most appropriate ordinal encoding for different types of categorical variables.In this article, we will continue…