Skip to main content
JH
Home
Experience
Projects
Lab
Resume
Blog
Live
>_
Back to Projects
Diabetes Prediction
Stacking ensemble combining 4 base learners with a meta-model.
Stacking
scikit-learn
Cross-Validation
IEEE
Ensemble Architecture
Stacking Architecture
2-Layer Ensemble
Logistic
78%
Random
80%
Gradient
79%
SVM
77%
Meta-Learner
Layer 1
Layer 2
Accuracy Comparison
+2.68% improvement
75%
80%
85%
78%
Logistic
80%
RF
79%
GBoost
77%
SVM
82.68%
Stack
Feature Importance
Glucose
26%
BMI
18%
Age
15%
Pregnancies
12%
Blood Pressure
10%
Insulin
8%
Skin Thickness
6%
Pedigree
5%
Training Pipeline
Preprocess
Scale + Impute
Base Models
4 Classifiers
Meta-Learner
Logistic Reg
Evaluate
5-Fold CV
82.68%
Accuracy
4
Base Models
5-Fold
Cross-Validation
8
Features