A new variable importance measure for random forests with missing data.
Alexander HapfelmeierTorsten HothornKurt UlmCarolin StroblPublished in: Stat. Comput. (2014)
Keyphrases
- missing data
- random forests
- random forest
- logistic regression
- missing values
- low rank
- ensemble methods
- machine learning algorithms
- decision trees
- incomplete data
- matrix factorization
- tree ensembles
- randomized trees
- decision tree ensembles
- support vector
- similarity measure
- shape analysis
- multiple imputation
- data mining
- relational databases
- image processing