Probability Calibration for Imbalanced Dataset

A suggestion to Undersampling method

Kyosuke Morita
Towards Data Science
8 min readNov 18, 2019

--

Photo by Bharathi Kannan on Unsplash

When we are trying to build a machine learning model for real-world problems, we are often faced with imbalanced datasets. Resampling method, especially undersampling is one of the most widely used methods to overcome the class imbalance (I also showed how those resampling methods worked on my Master dissertation in the…

--

--