Prediction of diabetes - Type 2 (Mellitus)

Hosted Link : Click here

This is a machine learning project based on the prediction of type 2 diabetes, with a given data. It uses logistic regression to classify the diabetic outcomes of each person's record.The diabetes is growing threat nowadays, one of the reasons being that there is no perfect cure for it. There are actually two types of diabetes, namely

• Type 1
• Type 2

The type 2 diabetes is commonly called diabetes mellitus. It can be defined as a chronic condition that affects the way the body processes blood sugar (glucose). We consider the mellitus here. After several researches we found that, some parameters are directly responsible for the mellitus to occur. By using the data of the people with diabetes and without diabetes, a dataset has been build. We use that dataset to classify the people who are in the risk of getting diabetes.

Directory structure

Code

This directory contains the source code files.

• Prediction.rmd : Source code,R Markdown file

Data

• Diabetes.csv : The dataset

Results

This directory has the codes and the outputs, presented in a neat format.

• Prediction.html : Presents the project as html file, with wonderful designs (Download is recommended)
• Prediction.pdf : Presents the project as simple document.

Table of contents

The topics are elaborated in the markdown/results

1. Introduction
2. Loading the required libraries
3. Data Loading
4. Correlations
5. Visualizations
6. Preparing the data
7. Logistic Regression
8. Prediction
9. ROC Curve
10. Comparison
11. Conclusion