Skip to main content

Table 6 Accuracy of TL for predicting the second level ATC code

From: Predicting anatomic therapeutic chemical classification codes using tiered learning

2 nd level ATC Code Prediction using different initializations and comparison to STP

Initialization method

Supervised learning

SuperPred

ChEMBL (Perfect 1st level)

STP

Prediction method

A(2)

CA(2)

(N(2))

A(2)

CA(2)

(N(2))

A(2)

CA(2)

(N(2))

A(2)

CA(2)

(N(2))

MLP

69.18%

71.42%

(64.34%)

70.79%

74.37%

(68.47%)

79.03%

89.52%

(79.03%)

69.89%

71.77%

(67.56%)

Naive Bayes

48.21%

47.94%

(36.74%)

60.22%

69.09%

(57.35%)

65.41%

82.71%

(65.05%)

46.95%

47.31%

(36.38%)

Random Forest

63.98%

65.23%

(57.71%)

68.28%

73.12%

(66.49%)

77.60%

88.80%

(77.60%)

62.19%

64.34%

(58.96%)

SVM

67.92%

69.89%

(63.26)

72.40%

75.18%

(70.79%)

81.90%

90.95%

(81.90%)

68.46%

70.16%

(64.52%)

  1. The best STP results are included for comparison. The highest prediction accuracies obtained with each initialization strategy are highlighted in bold