Module 02: Introduction to Machine Learning
Understand what machine learning is, how it works, and the different types of ML problems.
Before Module 04 (classification): Skim Ethics in ML (30–45 min). Module 04 introduces fairness tooling, ethical framing helps you use it well.
What You'll Learn
- What is Machine Learning?
- Types of Machine Learning
- ML Workflow and Best Practices
- Real-world Applications
ML for beginners curriculum map
| Topic | Where to study (examples in guide) |
|---|---|
| Machine learning for beginners | Guide |
| Data types in ML; population vs sample | Key concepts, Descriptive statistics and sampling |
| Descriptive statistics (mean, median, mode, variance, std) | Descriptive statistics and sampling |
| Types of ML; workflow; first project | Topics covered in this README; First ML project |
| Next: regression, classification, EDA, features | Regression, Classification, EDA, Feature engineering |
| Projects implementing algorithms | Beginner projects |
Topics Covered
What is Machine Learning?
- Definition and Core Concepts
- Difference from Traditional Programming
- Why Machine Learning Matters
- History and Evolution
Types of Machine Learning
1. Supervised Learning
- Learning from labeled data
- Regression vs Classification
- Examples: Predicting house prices, spam detection
2. Unsupervised Learning
- Learning from unlabeled data
- Clustering and Dimensionality Reduction
- Examples: Customer segmentation, anomaly detection
3. Reinforcement Learning
- Learning through interaction
- Reward-based learning
- Examples: Game playing, robotics
Machine Learning Workflow
- Problem Definition: Understand the business problem
- Data Collection: Gather relevant data
- Data Preparation: Clean and preprocess data
- Feature Engineering: Create meaningful features
- Model Selection: Choose appropriate algorithms
- Training: Train the model on data
- Evaluation: Assess model performance
- Deployment: Put model into production
- Monitoring: Track model performance over time
Best Practices
- Start with simple models
- Use appropriate evaluation metrics
- Avoid data leakage
- Split data properly (train/validation/test)
- Document everything
- Version control your code and data
Job-ready checkpoint (before Modules 03–05)
You are ready for supervised learning (regression and classification) when you can:
- Distinguish regression, classification, and clustering for real scenarios
- Sketch the ML workflow from business question to evaluation
- Complete the First ML Project Tutorial (Iris) without only copy-pasting
- Explain train/test split and why leakage invalidates metrics
- Start or continue a beginner portfolio project
Analyst / data scientist track: Begin Module 19 SQL in parallel if you have not already. See Foundation & Job Market Readiness.
Additional Resources
- ML Terminology Reference. Glossary of ML terms with examples
- Problem Identification & Algorithm Selection - Guide to identifying ML problems and choosing algorithms
- Your First ML Project Tutorial - Step-by-step walkthrough building an Iris classifier
- Common Pitfalls & Best Practices: Avoid mistakes and follow best practices
- AI Myths Busted: Wrong mental models about AI, ML, and deep learning (understanding, scale, RAG, agents, eval)
- Advanced Introduction to ML Topics - Advanced ML workflow patterns, system design principles, data-centric vs model-centric approaches, ML project management, and production considerations
- Introduction to ML Quick Reference - Quick lookup guide for algorithm selection, ML workflow checklist, problem type identification, terminology, and common patterns
Learning Objectives
By the end of this module, you should be able to:
- Explain what machine learning is and when to use it
- Distinguish between different types of ML (Supervised, Unsupervised, Reinforcement)
- Understand the complete ML workflow from problem to deployment
- Identify which type of ML to use for a given problem
- Select appropriate algorithms for different scenarios
- Build your first ML model from scratch
- Understand common ML terminology
- Avoid common pitfalls and follow best practices
Exercises
- Identify ML Problem Types: Classify real-world scenarios (regression/classification/clustering)
- Design ML Workflow: Create workflow for a given business problem
- Build First Model: Complete the First ML Project Tutorial
- Algorithm Selection: Use the Algorithm Selection Guide to choose algorithms for different problems
- Terminology Quiz: Test your understanding using the ML Terminology Reference
Key Concepts
- Training Data: Data used to teach the model
- Testing Data: Data used to evaluate the model
- Features: Input variables
- Labels/Targets: Output variables (for supervised learning)
- Model: The learned function that makes predictions
- Overfitting: Model memorizes training data, performs poorly on new data
- Underfitting: Model is too simple, can't capture patterns
Documentation & Learning Resources
Official Documentation:
Free Courses:
- Machine Learning Crash Course - Google: Free comprehensive course
- Introduction to Machine Learning - MIT: Free MIT course
- Machine Learning (Coursera - Andrew Ng): Free audit available
- Machine Learning with Python (IBM Coursera): Free audit available
Tutorials:
- Scikit-learn Tutorial
- Machine Learning Tutorial (W3Schools)
- ML Basics (Kaggle Learn): Free micro-course
Video Tutorials:
Books (Free):
- Introduction to Statistical Learning (Free PDF): Free comprehensive textbook
- Hands-On Machine Learning (O'Reilly): Check library access
- Pattern Recognition and Machine Learning (Bishop): Advanced but comprehensive
Interactive Learning:
- Kaggle Learn - Intro to Machine Learning: Free hands-on course
- Google's Machine Learning Crash Course: Free with exercises
- Fast.ai Practical Deep Learning: Free practical ML course
Tools and Platforms:
- Scikit-learn Algorithm Cheat Sheet: Visual algorithm selection guide
- Google Colab: Free Jupyter notebooks in the cloud
- Kaggle Notebooks: Free ML environment with datasets
Previous Module: 01-python-for-data-science
Next Module: 03-supervised-learning-regression