Skip to main content

Table 5 Important attribute scores according to the improved random forest model

From: Application of data mining for predicting hemodynamics instability during pheochromocytoma surgery

Attributes Importance Scores of random forest Attributes Importance Scores of random forest Attributes Relief-F
Mean Decrease Accuracy Mean Decrease Gini Weight of attribute
bmi 19.3095 bmi 16.688 ctvalue −2.8871
size 9.6143 size 12.2722 prevma −3.3417
asa 5.7061 prevma 9.5934 arrhythmia −4.4444
hypertension 4.6416 ctvalue 9.3884 age −4.6933
ctvalue 4.6293 age 8.0945 bmi −6.7295
prevma 1.3311 hypertension 2.9614 size −9.2895
preblood 0.8616 asa 2.3838 asa −13.3596
arrhythmia 0.2419 preblood 1.5554 preblood −19.3992
dm 0.0276 dm 1.2103 hypertension −22.7188
age −0.4422 arrhythmia 0.3524 dm −23.1212