Skip to main content

Advertisement

Table 4 Experimental comparison between the number of incorrectly classified instances found on the test sets by GP and Support Vector Machine with exponent 2 on unbalanced datasets.

From: A comparison of machine learning techniques for survival prediction in breast cancer

  5 yrs 7.5 yrs
  GP SVM-K2 GP SVM-K2
best 9 10 12 13
average (SEM) 15.04 (0.41) 17.84 (0.42) 21.18 (0.49) 20.7 (0.46)
  1. The datasets are defined by survival status at endpoints 5 and 7.5 years.