Load Dataset (Kaggle - Diabetes Dataset)

Check Dataset Shape and Info

Handle Missing Values

Map 'diabetic' Column to Binary (0: Non-Diabetic, 1: Diabetic)

Perform Summary Statistics

Identify Numerical and Categorical Columns

Label Encode Categorical Columns (Gender, Family History)

Split Data into Features (X) and Target (y)

Train-Test Split (80%-20%)

Scale Features (StandardScaler)

Class Imbalance Handling

Upsample Minority Class using SMOTE

Train Random Forest Model

Evaluate Model (Accuracy, Precision, Recall, F1-Score)

Train Logistic Regression Model

Evaluate Model (Accuracy, Precision, Recall, F1-Score)

Train SVM (Linear Kernel)

Evaluate Model (Accuracy, Precision, Recall, F1-Score)

Train Advanced SVM (Hyperparameter Tuning via GridSearch)

Evaluate Model (Accuracy, Precision, Recall, F1-Score)

Train XGBoost Model

Evaluate Model (Accuracy, Precision, Recall, F1-Score)

Apply Ensemble Learning (Soft Voting - XGBoost + RF)

Evaluate Ensemble Model (Accuracy, Precision, Recall, F1-Score)

Apply Ensemble Learning (Hard Voting - XGBoost + RF + LGBM)

Evaluate Ensemble Model (Accuracy, Precision, Recall, F1-Score)

Model Evaluation

Confusion Matrix and ROC-AUC Curve for Performance Analysis

Final Model Selection (Best Performing Model Based on Metrics)