Skip to main content

Table 8 To find the accuracy of each predictive model under different variable systems, a 5-fold-cross validation was used

From: Predicting postoperative complications of head and neck squamous cell carcinoma in elderly patients using random forest algorithm model

Algorithm variable

Accuracy

Support Vector Machine (SVM)

Random Forest (RF)

Rotation Forest (ROF)

Bayesian Network (BN)

Naïve Bayesian Network (NBN)

Alla

83.431 %

89.084 %

85.965 %

82.261 %

75.634 %

Newb

81.676 %

87.135 %

83.041 %

77.778 %

73.489 %

POSSUMb

79.337 %

82.261 %

79.142 %

74.074 %

71.929 %

APACHE IIb

75.439 %

76.420 %

76.023 %

73.294 %

75.829 %

  1. The predictive model based on the Random Forest algorithm has the best accuracy of 89.084 % under the “All variables” system
  2. aAll variables included in the study
  3. bVariables included in the model
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us