Table 4 Comparison of AUC values between baseline model (only with uncensored data) and weighting model in decision-tree modeling for classification problem.

From: A Machine Learning Approach Using Survival Statistics to Predict Graft Survival in Kidney Transplant Recipients: A Multicenter Cohort Study

Period after KT for prediction (year)

1 yr

2 yr

3 yr

4 yr

5 yr

6 yr

7 yr

8 yr

9 yr

10 yr

Baseline model AUCa,b (complete case analysis)

0.9728

0.8856

0.7916

0.7523

0.7121

0.7164

0.6707

0.6859

0.6714

0.6583

Weighting model AUCa,b

0.9754

0.8845

0.7958

0.7389

0.7013

0.720

0.7107

0.7150

0.7013

0.7066

  1. aSetting 1: Use overall attributes with multiple imputation, bInstance weighting methods by Zupan et al.6; yr, year.
Back to article page