Skip to content

This repository contains ML project on diabetes prediction in early stages.

License

Notifications You must be signed in to change notification settings

SuhasPK/diabetes-ML-app

Repository files navigation

Diabetes Prediction Web App

CLICK HERE! to check the web app.

Project Overview

This web application is designed to predict the risk of diabetes in individuals based on early-stage symptoms. The application is developed using Streamlit, a Python-based framework for building interactive web apps. The prediction model is powered by machine learning algorithms, specifically Logistic Regression and Decision Tree classifiers.

Author

Suhas. P. K

Data Source

The dataset used for training and testing the models is sourced from the Early Stage Diabetes Risk Prediction Dataset available on the UCI Machine Learning Repository. This dataset contains multiple health indicators that are used to predict the likelihood of diabetes.

Programming Language

The web app is built using Python, leveraging its robust libraries for machine learning and web development.

Machine Learning Models

Two different machine learning models have been implemented to predict the risk of diabetes:

  • Logistic Regression: A linear model used for binary classification, providing probabilities that an individual may develop diabetes.
  • Decision Tree: A non-linear model that creates a tree-like structure to make decisions based on the input features.

Key Features

  • User-Friendly Interface: Simple and intuitive interface for users to input their health data.
  • Real-Time Predictions: Instant prediction results based on user input.
  • Model Comparison: Display of prediction results from both Logistic Regression and Decision Tree models for comparison.

How to Use

  1. Input your health data into the form provided in the web app.
  2. Click on the "Predict" button to view the prediction results.
  3. Compare the results from both models to understand the risk of diabetes.

Future Enhancements

  • Integration of additional machine learning models for improved accuracy.
  • Inclusion of more health indicators to enhance prediction capability.
  • Deployment of the web app for public access.

Disclaimer: This web app is intended for educational purposes and should not be used as a substitute for professional medical advice.

About

This repository contains ML project on diabetes prediction in early stages.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published