Skip to content

Educational Jupyter notebook demonstrating how the five tribes of machine learning (from 'The Master Algorithm') approach the Iris classification problem

Notifications You must be signed in to change notification settings

codearranger/five-tribes-ml

Repository files navigation

Five Tribes Iris Classification

Open In Colab

An educational Jupyter notebook demonstrating how the five tribes of machine learning (from Pedro Domingos' "The Master Algorithm") approach the classic Iris flower classification problem.

Overview

This project compares five fundamental approaches to machine learning:

  • 🌳 Symbolists (Decision Trees) - Learning through logical rules
  • 🧠 Connectionists (Neural Networks) - Learning by mimicking the brain
  • 🧬 Evolutionaries (Genetic Algorithms) - Learning through simulated evolution
  • 📊 Bayesians (Naive Bayes) - Learning through probabilistic inference
  • 📏 Analogizers (k-Nearest Neighbors) - Learning by recognizing similarity

Installation

# Clone the repository
git clone <your-repo-url>
cd masteralgorithm

# Install dependencies
pip install -r requirements.txt

Usage

# Start Jupyter
jupyter notebook

# Open: five_tribes_iris_classification.ipynb

Or view the pre-executed notebook directly on GitHub!

What You'll Learn

  • The philosophical differences between ML paradigms
  • When to use each type of algorithm
  • Working implementations you can modify and experiment with
  • How the same problem can be solved in fundamentally different ways

Requirements

  • Python 3.9+
  • See requirements.txt for package dependencies

Inspiration

This project is inspired by Pedro Domingos' book "The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World".

License

MIT License - feel free to use this for learning and teaching!

Contributing

Suggestions and improvements welcome! Please open an issue or pull request.

About

Educational Jupyter notebook demonstrating how the five tribes of machine learning (from 'The Master Algorithm') approach the Iris classification problem

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •