The authors caution that KDDCUP99 is inherently oversampled: using only 50 randomly sampled training examples, either an SVM or least squares with random features consistently produced test error on the order of 8%. · CiteArk
The authors caution that KDDCUP99 is inherently oversampled: using only 50 randomly sampled training examples, either an SVM or least squares with random features consistently produced test error on the order of 8%.
