Skip to content

This project features a classification model built using the PIMA Indian Diabetes dataset and showcases the use of data visualisation techniques to identify outliers and cleanse the data for an increased accuracy.

Notifications You must be signed in to change notification settings

NotHari/Diabetes-and-Pre-Diabetes-Predictor

Repository files navigation

Prediction of Pre-Diabetes and Diabetes in Mothers above 21 years of age with Machine Learning Models

image

The purpose of this project titled, ‘Prediction of Pre-Diabetes and Diabetes in Mothers above 21 years of age with Machine Learning Models’, was to build a website which collects relevant user details and predicts their diabetic condition using a Machine Learning model.

Table of Contents

About the Project

The project titled 'Prediction of Pre-Diabetes and Diabetes in Mothers above 21 years of Age' features several data visualisation and pre-processing techniques before feeding the data to various Machine Learning Models to obtain the best trained model which is then used to predict diabetic condition of a patient via the website.

The target users for our project are mothers over the age of 21 who are relatively more prone to diabetes. They can use the website for self-diagnosis, helping them make better health decisions and get medical consultation to avoid future complications.

Library Used

The project made use of libraries like Matplotlib to visualise the data and observe patterns and outliers and facilitated the data cleansing process. Scikit-Learn was used to implement the Machine Learning Models.

Flask was used to deploy the model and make predictions after inputting the various health attributes.

Interface

The below image shows the website interface that makes predictions using the trained Machine Learning Model.

Results

From the analysis, the best performing model was the Random Forest Classifier with an increased accuracy of 81%.

Made with ❤

About

This project features a classification model built using the PIMA Indian Diabetes dataset and showcases the use of data visualisation techniques to identify outliers and cleanse the data for an increased accuracy.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •