Reducing the Overfitting of Adaboost by Controlling its Data Distribution Skewness.
Yijun SunSinisa TodorovicJian LiPublished in: Int. J. Pattern Recognit. Artif. Intell. (2006)
Keyphrases
- data distribution
- decision trees
- data streams
- object detection
- training error
- learning algorithm
- distributed data
- communication cost
- index structure
- data points
- boosting algorithms
- high dimensional data
- face detection
- multi class
- support vector
- training instances
- adaboost algorithm
- decision boundary
- base classifiers
- streaming data
- concept drift
- generalization error
- cross validation
- ensemble learning
- active learning
- training data
- multi dimensional data
- data skew