Heart Disease Analysis (Classification Approaches)
Authors: Ishan V Mahajan
Abstract: The dataset used in this project is the Heart Disease dataset from the UCI Machine Learning Repository. It contains various medical attributes related to heart disease diagnosis.
Source: UCI Machine Learning Repository
This project aims to analyze the Heart Disease dataset from the UCI Machine Learning Repository. The primary objectives are to perform classification using logistic regression and XGBoost, compare these results with Bayesian modeling using Bernoulli distributions, and present the findings in a Python Shiny app.
Full Description: - Results
The results of the analysis are presented in the Python Shiny app, which includes:
Distribution of key factors
Logistic regression model fitting
Odds of heart disease
Bayesian Probabilities
Shiny App
The Shiny app provides an interactive interface to explore the analysis results. It includes various panels for different aspects of the analysis:
Distributions: Visualize the distribution of key factors.
Heart DIsease: View logistic regression model fitting.
Odds of having Heart Disease: Summarize the odds of having heart disease.
Bayesian Approach: Demonstrate the Bayesian posterior probabilities of factors causing Heart Disease.
Future Aim: Build Robust Classification models and compare them to Bayesian Approaches. Build this app into a general purpose app that compares Frequentist Vs Bayesian Classification approaches.
Shiny app: Understanding Heart Disease
Repo: GitHub - ivm25/likelihood_of_an_event_using_py_shiny: Comparing frequentist and bayesian approaches to understand the odds of heart disease
Thumbnail:
Full image: