Asthma Prediction Project

Overview

This project demonstrates a complete end-to-end data science pipeline using the Asthma Disease Dataset.
The goal is to predict which factors lead to a positive asthma diagnosis, following the typical data science lifecycle:

1. Data exploration & visualization
2. Feature engineering
3. Model building & evaluation
4. Packaging reusable code
5. Testing and documentation

This project is part of the SoftUni Data Science course.

---

Project Structure

asthma_project/
│
├── data/               # Dataset(s)
│   └── asthma_disease_data.csv
│
├── notebooks/          # Jupyter notebooks (EDA, experiments)
│   ├── 01_exploration.ipynb
│   ├── 02_feature_engineering.ipynb
│   └── 03_modeling.ipynb
│
├── src/                # Python source code (reusable functions)
│   ├── __init__.py
│   ├── data_prep.py
│   ├── features.py
│   └── model.py
│
├── tests/              # Unit tests
│   └── test_data_prep.py
│
├── outputs/            # Plots, results, model artifacts
│
├── requirements.txt    # Dependencies
├── README.md           # Project description
└── .gitignore          # Ignore rules

---

Dataset

The dataset is provided by the course and is located in data/asthma_disease_data.csv.
It contains patient health data and labels indicating asthma diagnosis.

---

Setup

1. Clone the repository (private now):

   git clone https://github.com/YOUR-USERNAME/asthma-prediction.git
   cd asthma-prediction

2. Create a virtual environment and install dependencies:

   python -m venv venv
   venv\Scripts\activate # On Windows
   pip install -r requirements.txt

3. Launch Jupyter:

   jupyter lab

4. Tests:

   python -m pytest -v

---

Goals

• Explore dataset properties and key patterns.
• Engineer useful features for prediction.
• Train and evaluate classification models.
• Package workflow into reusable modules (src/).
• Add unit tests for reproducibility.

---

Notes

• This project is for educational purposes as part of the SoftUni Data Science course.
• The workflow is simplified compared to real-world projects, which are more iterative and complex.