Knowledge Base
Clear, practical guides to the data science methods we use every day — written for business-minded people who want to understand the technology, not just trust it blindly.
All Topics
How to model the relationship between variables and make quantitative predictions — from simple lines of best fit to full multiple regression with validation.
Decision trees, random forests, and gradient boosting — the workhorses of applied ML. Learn how splitting rules and ensemble methods create powerful predictors.
From perceptrons to deep networks — how layers of interconnected nodes learn complex patterns, and why activation functions and backpropagation matter.
What LLMs actually are, how transformers and attention work, and how to think about integrating them into real business workflows without the hype.
K-Means, DBSCAN, hierarchical clustering — how to find natural groupings in data without labels, and when each approach makes sense for your problem.
PCA, t-SNE, UMAP — techniques for compressing high-dimensional data into something visualisable and useful, without losing the signal that matters.