Decision trees for binary classification variables grow equally with the Gini impurity measure and Pearson's chi-square test.
Johannes GrabmeierLarry A. LambePublished in: Int. J. Bus. Intell. Data Min. (2007)
Keyphrases
- binary classification
- chi square test
- decision trees
- ensemble methods
- logistic regression
- support vector
- multi class
- learning problems
- cost sensitive
- multi class classification
- support vector machine
- naive bayes
- machine learning algorithms
- multi label
- correlation coefficient
- machine learning
- odds ratio
- prediction accuracy
- generalization error
- training data
- training set
- similarity measure
- class imbalance
- feature selection
- base classifiers
- image segmentation
- learning tasks
- cross validation
- markov random field