This repository contains the code base and all the necessary files to implement the diabetes prediction system through machine learning via a webapp.
In this project we attempt to implement a machine learning approach to predict the probability of a patient having diabetes based upon his reports. Predicting the chances of a person having Diabetes can be done effectively using Machine Learning.
The main objective is to predict the chances of a person having Diabetes so that effective treatment can be provided to him/her. Our proposed Diabetes Prediction System integrates mathematical functions, machine learning, and training models which can help to understand the effectiveness of the model. This can be used for the purpose of achieving better diabetes prediction accuracy and providing real life accurate results.
In our Diabetes Prediction Model, we have implemented an approach of developing our machine learning model based upon four models i.e. K-Nearest Neighbours, Logistic Regression, Support Vector Class and Random Forest Classifier. We have taken our dataset, which consists of values such as BMI, Insulin and Glucose levels, Age, Blood Pressure etc. and split it into train and test data. Then we will perform steps like understanding and identifying data, cleanse the data, Exploratory Data Analysis, Train and Test the model based upon train and test data split. We will find the accuracy of the all models and will select the best one for our diabetes prediction. At last, we will provide patient’s details as input for the model which will provide the probability of him/her having diabetes.
The choice of development environment of our project is ‘Jupyter Notebook’. Jupyter Notebook is an online cloud based versatile environment and supports various libraries for python programming, machine learning and deep learning.
Also, we have implemented the front-end user input portal via HTML, CSS and JavaScript to connect the respective files to the ML model via Python Flask, a web framework in python to develop and connect python applications to web applications.