A comprehensive machine learning project for predicting heart disease using the UCI Heart Disease Dataset. This project combines data analysis, visualization, and multiple machine learning algorithms to create an accurate heart disease prediction system with an interactive web interface.
🌟 Features
📊 Comprehensive Data Analysis: In-depth exploratory data analysis with statistical insights
🎯 Multiple ML Models: Comparison of Random Forest, Logistic Regression, SVM, and KNN algorithms
🚀 Interactive Web App: Real-time predictions through a user-friendly Streamlit interface
📈 Rich Visualizations: Interactive charts and graphs using Plotly
🔍 Feature Analysis: Correlation analysis and feature importance visualization
⚡ Real-time Predictions: Instant heart disease risk assessment
🤖 Machine Learning Models
Model Performance Comparison
Model | Accuracy | Strengths |
|---|---|---|
🔍 K-Nearest Neighbors | 91.8% | Simple, effective for local patterns, excellent performance |
🎯 Support Vector Machine | 90.2% | Good with non-linear data, robust |
🌳 Random Forest | 86.9% | Handles non-linear relationships, good interpretability |
📈 Logistic Regression | 85.5% | High interpretability, fast training |
