Health Intelligence Platform

A healthcare prediction system built progressively: ML → Statistics → Database → Big Data

Table of Contents

• Overview
• Features
• Prerequisites
• Installation
• Getting Started
• Tech Stack
• Project Structure
• Documentation
• Resources

---

Overview

Stage	Course	Focus
1	ESE 5410	Machine Learning
2	ESE 5420	Statistics
3	CIS 5500	Database
4	CIS 5450	Big Data

Development Focus: Stage 1 (Machine Learning)

See ROADMAP.md for timelines and detailed deliverables per stage.
See PROJECT.md for use case and design principles.

---

Features

Stage 1 (Current) — Diabetes Risk Prediction

• Load and preprocess healthcare data
• Build multiple ML models (logistic regression, random forest, SVM, ensemble, neural networks)
• Evaluate models with cross-validation
• Feature importance analysis
• Uncertainty quantification

For complete Stage 1 deliverables, see ROADMAP.md.

Future Stages

• Statistical validation (Stage 2)
• Database-backed application (Stage 3)
• Distributed processing at scale (Stage 4)

---

Prerequisites

• Python 3.10+
• pip
• Git

---

Installation

1. Clone Repository

git clone https://github.com/yourusername/health-intelligence-platform.git
cd health-intelligence-platform

2. Create Virtual Environment

Windows:

python -m venv .venv
.venv\Scripts\activate

macOS/Linux:

python -m venv .venv
source .venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

4. Verify Installation

pytest tests/ -v

---

Getting Started

1. Select a healthcare dataset with binary outcome (500+ rows)
2. Place raw data in data/raw/
3. Start the notebook:

jupyter notebook notebooks/01_initial_model.ipynb

4. Follow TODO comments to build baseline model

See ROADMAP.md for complete stage breakdown.

---

Project Structure

src/          Reusable code (data, features, models, evaluate)
notebooks/    Jupyter notebooks (exploratory analysis)
data/         Raw and processed datasets
tests/        Unit tests

---

Tech Stack

Stage 1 (Current)

• Language: Python 3.10+
• ML & Data: pandas, NumPy, scikit-learn
• Notebooks: Jupyter
• Testing: pytest
• Visualization: Matplotlib, Seaborn

Future Stages

• Stage 2: scipy, statsmodels
• Stage 3: PostgreSQL, Flask/FastAPI
• Stage 4: Apache Spark, cloud platforms

For full technical decisions and rationale per stage, see ARCHITECTURE.md.

---

Documentation

Document	Purpose
ROADMAP.md	Stage definitions and deliverables
PROJECT.md	Use case, principles, constraints
COURSE_ALIGNMENT.md	How courses map to features
ARCHITECTURE.md	Technical decisions per stage
DOCS_STRUCTURE.md	How documentation is organized

---

Resources

Datasets

• Kaggle Datasets — Find healthcare datasets
• UCI Machine Learning Repository — Classic datasets
• Awesome Healthcare Datasets — Curated list

Learning

• scikit-learn Documentation
• Pandas Documentation
• An Introduction to Statistical Learning — Free textbook