== K-MEANS ==
k terpilih   : 2
Pencarian k:
  inertia    : 2=8497.58, 3=5886.21, 4=4408.92, 5=3764.81, 6=3153.72, 7=2684.73, 8=2283.15
  silhouette : 2=0.437, 3=0.360, 4=0.358, 5=0.370, 6=0.353, 7=0.372, 8=0.385
  calinski   : 2=2375.1, 3=2683.1, 4=2875.4, 5=2711.7, 6=2758.4, 7=2826.7, 8=2958.0

Centroid per cluster:
 cluster  recency_days  log_frequency  log_monetary  frequency  monetary
       0    696.251001       1.399452     14.339863       3.05 1689364.0
       1   1204.144632       0.716840     12.507634       1.05  270392.0

Profil per cluster:
 cluster    n  recency  frequency     monetary
       0  749    540.0       3.42 1782360011.0
       1 3623   1318.0       1.06 1894277225.0

== RANDOM FOREST ==
Akurasi (test)    : 0.9991
Cross-val (5-fold): [0.6526, 0.9360, 0.9908, 0.9588, 0.8318] | mean=0.8740

Classification report:
precision    recall  f1-score   support

           0       1.00      0.99      1.00       187
           1       1.00      1.00      1.00       906

    accuracy                           1.00      1093
   macro avg       1.00      1.00      1.00      1093
weighted avg       1.00      1.00      1.00      1093

Feature importance:
     feature  gini_importance  permutation_importance
   frequency         0.579002                0.204392
    monetary         0.290738                0.074565
recency_days         0.130260                0.039250