A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection
A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection
Ron Kohavi
1995 · DBLP: conf/ijcai/Kohavi95
International Joint Conference on Artificial Intelligence · 引用 13,395 次
TLDR
The results indicate that for real-word datasets similar to the authors', the best method to use for model selection is ten fold stratified cross validation even if computation power allows using more folds.