This project aims to predict diabetes using the XGBoost classifier, leveraging the Pima Indians Diabetes dataset from Kaggle, which contains essential features like glucose levels, BMI, age, insulin levels, and more. The goal is to build a predictive model that can accurately classify individuals as diabetic or non-diabetic based on these medical attributes. XGBoost, known for its efficiency and effectiveness in handling large datasets with complex relationships, is employed to train the model. Data preprocessing steps, including handling missing values and feature scaling, are applied to optimize the model's performance. Hyperparameter tuning is conducted to find the best model configuration, enhancing accuracy and minimizing overfitting. The model is evaluated using standard performance metrics such as accuracy, precision, recall, and F1-score, ensuring it is robust and reliable. Feature importance analysis is performed to identify which attributes, such as glucose levels and BMI, have the greatest influence on the prediction, providing insights into the key factors contributing to diabetes. The XGBoost classifier demonstrates strong performance with high accuracy, precision, and recall, making it a powerful tool for early diabetes detection.

This is a Diabetes Prediction web application built using HTML, CSS, and the Django framework.
Users can input key medical parameters such as Pregnancies, Glucose Level, Blood Pressure, BMI, and more into a clean, responsive form.

The application processes the input and predicts the likelihood of diabetes based on trained machine learning models on the backend.

Django manages the form submission, model integration, and prediction logic, while HTML and CSS ensure a modern and user-friendly interface.

The goal is to provide a simple, fast, and accessible tool to help users assess their diabetes risk.