The Unsupervised Learning Workshop: Get started with unsupervised learning algorithms and simplify your unorganized data to help make future predictions

Unsupervised learning lives at the heart of modern data exploration—finding patterns, simplifying complexity, and preparing raw data for better predictions. In The Unsupervised Learning Workshop, Aaron Jones, Christopher Kruger, and Benjamin Johnston turn a broad field into a practical, exercise-driven path for Python users. 📊

What Makes This Workshop Different

Instead of a theory-only overview, each chapter pairs clear explanations with hands-on coding exercises and activities. You will work through clustering, dimensionality reduction, autoencoders, t-SNE visualizations, topic modeling, and market basket analysis using popular Python libraries such as scikit-learn, Keras, pandas, and Matplotlib.

Algorithms You’ll Explore

• Clustering: k-means, hierarchical clustering, and DBSCAN, including distance metrics and evaluation with silhouette scores.
• Dimensionality Reduction: PCA from scratch and with scikit-learn, eigenvalue intuition, and variance visualization.
• Autoencoders: neural network fundamentals, simple and multi-layer autoencoders, and convolutional autoencoders with Keras.
• t-SNE: building meaningful low-dimensional visualizations and tuning perplexity and iterations.
• Topic Modeling: LDA with variational inference and NMF with TF-IDF, plus data cleaning for text.
• Market Basket Analysis: probabilistic metrics and the Apriori algorithm on transaction data.

Learn by Building and Experimenting

Every chapter guides you through small exercises and larger activities that let you apply new concepts immediately. The structure encourages experimentation rather than passive reading, building confidence in both the algorithms and the Python implementations.

Who Should Pick This Up

The workshop is especially suited for data analysts, aspiring data scientists, and developers with basic Python experience who want a practical introduction to unsupervised learning. It provides enough mathematical intuition to understand what the algorithms do without becoming a heavy theory textbook.

By the end of the workshop, you will have working code and a clearer mental map of how unsupervised methods unlock structure in unlabeled data.