The Data Science Workshop: Learn How You Can Build Machine Learning Models and Create Your Own Real-World Data Science Projects

A Workshop That Puts Python to Work

Data science can feel like a collection of scattered techniques until you see how the pieces fit together. This Second Edition takes a deliberately practical route: instead of drowning you in theory, it walks through real datasets, clear exercises, and activities that mirror the kind of work data scientists do every day. The result is a learning experience that feels closer to a mentor-led workshop than a dry reference manual.

What You’ll Build Along the Way

The book moves through the key stages of a data science workflow. You’ll start with loading and preparing data in Python, then move into regression analysis to understand relationships between variables. From there, the material tackles binary and multiclass classification problems, cluster analysis for segmentation, and the critical task of measuring how well a model performs.

• Regression: Simple and multiple linear regression, log-linear models, and using the statsmodels formula API.
• Classification: Logistic regression for binary problems and random forests for multiclass tasks, with tuning techniques to reduce overfitting.
• Clustering: k-means clustering, choosing the right number of clusters, and standardizing data for better results.
• Evaluation: Metrics such as R², MAE, precision, recall, F1 score, confusion matrices, ROC curves, and AUC.

Hands-On Exercises and Real Data

Each chapter includes step-by-step exercises and end-of-chapter activities that encourage you to apply what you’ve learned. You’ll work with datasets related to banking, animal classification, Australian postcodes, and more. The book also introduces essential Python libraries like scikit-learn and statsmodels, giving you a foundation you can reuse in your own projects.

Why This Workshop Approach Helps

By working through concrete examples, you’ll develop an intuition for when to use a particular model, how to tune it, and how to interpret results. The authors don’t just show you code; they explain the reasoning behind each step, so you learn to make better decisions when you face new problems. This is particularly valuable if you’ve completed an introductory Python course but still feel unsure about building models from scratch.

Who Should Read This Book

If you’re a student, analyst, or developer who wants to move beyond basic Python and start building real machine learning models, this book provides a structured path. It assumes some familiarity with Python but does not expect a deep background in statistics or machine learning. The clear explanations and progressive exercises make it suitable for self-study or as a supplement to a data science course.

From Concepts to Production-Ready Skills

The Second Edition has been updated to reflect current best practices and includes additional guidance on model evaluation and tuning. By the end of the book, you’ll have a portfolio of practical skills: preparing data, building regression and classification models, clustering data, and assessing performance—all essential steps toward creating your own data science projects.