Skip to main content

Table 4 Descriptive Statistics of AUC for Different Models

From: Study on risk factors of diabetic peripheral neuropathy and establishment of a prediction model by machine learning

 

Min

Q1

Median(Q2)

Q3

Max

P value

XGBoost

0·679

0·709

0·764

0·775

0·801

0·074

RF

0·686

0·732

0·736

0·748

0·765

0·203

LR

0·586

0·675

0·683

0·720

0·737

0·040

KNN

0·607

0·650

0·671

0·698

0·739

0·853

NB

0·543

0·568

0·589

0.614

0·634

0·547

DT

0·636

0·663

0·679

0·690

0·759

0·299

  1. P < 0.05 are in bold
  2. Performing a Shapiro-Wilk test for normality on the AUC values from 10-fold cross-validation
  3. XGBoost Extreme Gradient Boosting, NB Naive Bayes, LR Logistic Regression, KNN K-Nearest-Neighbor, RF Random Forest, DT Decision Tree
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us