This project implements a full end-to-end machine learning pipeline to predict the presence of heart disease based on the UCI Heart Disease Dataset. The workflow covers data preprocessing, feature selection , model training, hyperparameter tuning, and deployment of an interactive web application using Streamlit and Ngrok.
- Data Preprocessing: Handles missing values, encodes categorical features, and scales numerical data.
- Dimensionality Reduction: Applies Principal Component Analysis (PCA) to visualize data structure.
- Feature Selection: Utilizes Random Forest Importance, RFE, and Chi-Square tests to identify the most relevant predictors.
- Supervised Learning: Trains and evaluates multiple classification models, including Logistic Regression, Decision Trees, Random Forest, and SVM.
- Unsupervised Learning: Employs K-Means and Hierarchical Clustering to discover natural groupings in the data.
- Model Optimization: Fine-tunes the best-performing model using GridSearchCV for enhanced accuracy.
- Interactive UI: A user-friendly web interface built with Streamlit for real-time predictions.
- Deployment: Instructions for making the local web application publicly accessible via Ngrok.
This project uses the Heart Disease UCI Dataset. It is a widely-used dataset for classification tasks, containing 76 attributes, but only 14 are typically used for analysis.
- Link to Dataset: UCI Machine Learning Repository: Heart Disease Data Set
Heart_Disease_Project/
├── data/
│ └── heart_disease.csv
├── deployment/
│ └── ngrok_setup.txt
├── models/
│ └── final_model.pkl
├── notebooks/
│ ├── 01_data_preprocessing.ipynb
│ ├── 02_pca_analysis.ipynb
│ ├── 03_feature_selection.ipynb
│ ├── 04_supervised_learning.ipynb
│ ├── 05_unsupervised_learning.ipynb
│ └── 06_hyperparameter_tuning.ipynb
├── results/
│ └── evaluation_metrics.txt
├── ui/
│ └── app.py
├── .gitignore
├── README.md
└── requirements.txt