Heart Disease Early Detection ML Pipeline

Overview

This project implements a machine learning pipeline for the early detection of heart diseases using Apache Spark, MLlib library, and a variety of machine learning algorithms including RandomForest, Logistic Regression, and XGBoost. The pipeline is designed to process a CSV database, apply transformations, train models, evaluate their performance, and showcase results through data visualization.

Project Highlights

• Machine Learning Algorithms:

  • RandomForest
  • Logistic Regression
  • XGBoost

• Processing Chain:

  • Transformers: Utilized 2 transformers in the processing chain.
  • Estimator: Trained models using various algorithms.
  • Pipeline: Structured a pipeline for seamless data processing.
  • Evaluation: Determined key metrics to assess model performance.

• Activity:

  • Hunted down a suitable CSV database for the project. https://www.kaggle.com/datasets/johnsmith88/heart-disease-dataset
  • Illustrated the processing chain with a focus on transformers, estimators, pipelines, and evaluation.
  • Programmed a notebook, rigorously tested, and ensured functionality.
  • Showcased results through effective data visualization.

Results

• Witness the power of data-driven insights!
• The pipeline demonstrated promising outcomes in the early detection of heart diseases.
• Achieved accuracy and efficiency through strategic algorithmic choices.

Usage

Prerequisites

• Python (version 3.11.5)
• Apache Spark