Skip to main content

Table 1 Comparison of classification performances on MAQC-II data set

From: A filter-based feature selection approach for identifying potential biomarkers for lung cancer

  Classification Algorithms
Feature Selection Methods Support Vector Machine k-Nearest Neighbor Naive Bayes Random Forest
Information Gain 0.9031 (6) 0.9380 (25) 0.9008 (40) 0.9206 (50)
Chi-squared test 0.8821 (1) 0.9164 (50) 0.9151 (4) 0.9441 (60)
Relief-F 0.8821 (1) 0.9052 (15) 0.8995 (50) 0.9306 (60)
t-test 0.9067 (15) 0.9100 (20) 0.9042 (8) 0.9304 (40)
Window t-test 0.8903 (5) 0.9216 (5) 0.9012 (2) 0.9199 (10)
Moderated t-test 0.8903 (6) 0.9084 (5) 0.8987 (1) 0.9309 (50)
BMI 0.9077 (4) 0.9298 (15) 0.9164 (4) 0.9250 (9)
  1. Each value represents the maximum AUC value (by 10-fold cross-validation) achieved by the corresponding feature selection method and classification algorithm. The number of features used to achieve the maximum is shown inside parenthesis.