Podcasts Technology Linear Digressions SMOTE: makin' yourself some fake minority data

SMOTE: makin' yourself some fake minority data

Published: June 13, 2016, 3:06 a.m.

Machine learning on imbalanced classes: surprisingly tricky. Many (most?) algorithms tend to just assign the majority class label to all the data and call it a day. SMOTE is an algorithm for manufacturing new minority class examples for yourself, to help your algorithm better identify them in the wild.\n\nRelevant links:\nhttps://www.jair.org/media/953/live-953-2037-jair.pdf