Diabetes Prediction

A Flask web app to predict diabetes in a patient using SVM ML model.

Table of contents

• About
• Experimental Setup
• Screenshots

About

• Diabetes can be controlled if it is predicted earlier. Hence, this project aims to perform early prediction of Diabetes in a patient by applying various Machine Learning Techniques.
• These techniques provide better results for prediction by constructing models from datasets containing various information about different people.
• Algorithms used were: K-Nearest Neighbor (KNN), Logistic Regression (LR), Decision Tree (DT), Support Vector Machine (SVM), Naïve Bayes(NB), and Random Forest (RF).
• The accuracy for each model was calculated.
• Results showed that SVM achieved higher accuracy compared to other machine learning techniques hence, it is used for prediction in the web application.

Experimental Setup

• Environment used:
  • Web App: Visual Studio Code
  • Model Training: Jupyter Notebook
• Languages & Libraries used:
  • Web App:
    • Front-end: HTML5, CSS3, Bootstrap v4.5
    • Back-end: Flask v1.1.2
  • Model Training:
    • Language: Python v3.8
    • Libraries: pandas v1.3.2, numpy v1.19.0, seaborn v0.11.2, matplotlib v3.4.3, scikit_learn v0.24.2

Screenshots