Unsupervised Algorithms- ML

Machine Learning Algorithms — Supervised Algorithms

Every ML model you'll ever build runs on an algorithm making decisions underneath. Knowing which algorithm to choose, why it behaves the way it does, and where it breaks is what separates someone who follows tutorials from someone who actually solves problems.

This resource covers the full landscape — not as a theoretical textbook, but as a practical guide to understanding, choosing, and applying the right algorithm for the right problem.

What's covered:

Unsupervised Learning

1. Principal Component Analysis — reducing dimensions without losing what matters
2. Anomaly Detection — finding what doesn't belong before it becomes a problem

Clustering

1. K-Means, DBSCAN, and Hierarchical Clustering — grouping data when you don't have labels
2. How to evaluate clusters when there's no ground truth to compare against

Cheatsheets included:

1. Algorithm selection guide — which algorithm for which problem, at a glance
2. Hyperparameter quick reference — what to tune and in which direction
3. Model evaluation metrics — accuracy, precision, recall, F1, RMSE, and when each one matters

Who this is for:

Anyone in data or ML who wants to stop picking algorithms by trial and error and start making deliberate, informed choices — whether you're preparing for interviews, building your first production model, or trying to explain your approach to a stakeholder.

The best model isn't the most complex one. It's the one you understood well enough to choose on purpose.