article thumbnail

ML internals: Synthetic Minority Oversampling (SMOTE) Technique

Domino Data Lab

note that this variant “performs worse than plain under-sampling based on AUC” when tested on the Adult dataset (Dua & Graff, 2017). Data mining for direct marketing: Problems and solutions. Proceedings of the Fourth International Conference on Knowledge Discovery and Data Mining, 73–79.

article thumbnail

Changing assignment weights with time-based confounders

The Unofficial Google Data Science Blog

Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining. Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2017. [4] 2] Scott, Steven L. 2015): 37-45. [3] 3] Hill, Daniel N., 2015): 37-45.

article thumbnail

Explaining black-box models using attribute importance, PDPs, and LIME

Domino Data Lab

Instead, you should focus on how techniques like PDPs and LIME can be used to gain insights into the model’s inner workings and how you can add those to your data science toolbox. In IJCAI 2017 Workshop on Explainable Artificial Intelligence (XAI), pages 24–30, Melbourne, Australia, 2017. References. Explainable planning.

Modeling 139