This project implements a machine learning model to predict the likelihood of diabetes based on patient data. The model is deployed as a web application using Flask, allowing users to input their health information and receive real-time predictions.
Exploratory data analysis (EDA) is performed to gain insights into the dataset. Visualizations include distplots, box plots, and heatmaps to analyze the distribution, outliers, and correlations among features.
Various machine learning algorithms are trained and evaluated to predict diabetes. The Random Forest classifier is selected as the best-performing model based on accuracy.
The trained model is integrated into a Flask web application. Users can input their age, BMI, HbA1c level, blood glucose level, gender, smoking history, hypertension, and heart disease status to get predictions on diabetes risk.
Dash and Plotly are utilized to create interactive visualizations, allowing users to explore relationships between different health conditions and diabetes.