Improving Decision Tree with AdaBoost

Diagnosis of Diabetes — Final Thesis (Skripsi)

Overview

This project evaluates how AdaBoost improves Decision Tree performance for diabetes diagnosis. It covers data preprocessing, model training, evaluation, and comparison with baseline models.

Highlights

Dataset: Pima Indians Diabetes
Methods: Decision Tree (baseline) vs. AdaBoost + Decision Tree
Metrics: Accuracy, Precision, Recall, F1-score

Resources

Notebook (train_test.ipynb)
Dataset (diabetes.csv)
README (raw)

Results (Summary)

Using AdaBoost with Decision Trees improved overall accuracy and recall compared to a standalone Decision Tree on the diabetes dataset. See the notebook for code and detailed metrics.