Numsense! Data Science for the Layman (No Math Added)

Description

Want to get started on data science? Our promise: no math added. This book has been written in layman’s terms as a gentle introduction to data science and its algorithms. Each algorithm has its own dedicated chapter that explains how it works, and shows an example of a real-world application. To help you grasp key concepts, we stick to intuitive explanations and visuals. Popular concepts covered include: - A/B Testing - Anomaly Detection - Association Rules - Clustering - Decision Trees and Random Forests - Regression Analysis - Social Network Analysis - Neural Networks Features: - Intuitive explanations and visuals - Real-world applications to illustrate each algorithm - Point summaries at the end of each chapter - Reference sheets comparing the pros and cons of algorithms - Glossary list of commonly-used terms With this book, we hope to give you a practical understanding of data science, so that you, too, can leverage its strengths in making better decisions. Foreword Preface Why Data Science? 1. Basics in a Nutshell 1.1 Data Preparation 1.2 Algorithm Selection 1.3 Parameter Tuning 1.4 Evaluating Results 1.5 Summary 2. k-Means Clustering 2.1 Finding Customer Clusters 2.2 Example: Personality Profiles of Movie Fans 2.3 Defining Clusters 2.4 Limitations 2.5 Summary 3. Principal Component Analysis 3.1 Exploring Nutritional Content of Food 3.2 Principal Components 3.3 Example: Analyzing Food Groups 3.4 Limitations 3.5 Summary 4. Association Rules 4.1 Discovering Purchasing Patterns 4.2 Support, Confidence and Lift 4.3 Example: Transacting Grocery Sales 4.4 Apriori Principle 4.5 Limitations 4.6 Summary 5. Social Network Analysis 5.1 Mapping out Relationships 5.2 Example: Geopolitics in Weapons Trade 5.3 Louvain Method 5.4 PageRank Algorithm 5.5 Limitations 5.6 Summary 6. Regression Analysis 6.1 Deriving a Trend Line 6.2 Example: Predicting House Prices 6.3 Gradient Descent 6.4 Regression Coefficients 6.5 Correlation Coefficients 6.6 Limitations 6.7 Summary 7. k-Nearest Neighbors and Anomaly Detection 7.1 Food Forensics 7.2 Birds of a Feather Flock Together 7.3 Example: Distilling Differences in Wine 7.4 Anomaly Detection 7.5 Limitations 7.6 Summary 8. Support Vector Machine 8.1 “No” or “Oh No”? 8.2 Example: Predicting Heart Disease 8.3 Delineating an Optimal Boundary 8.4 Limitations 8.5 Summary 9. Decision Tree 9.1 Predicting Survival in a Disaster 9.2 Example: Escaping from the Titanic 9.3 Generating a Decision Tree 9.4 Limitations 9.5 Summary 10. Random Forests 10.1 Wisdom of the Crowd 10.2 Example: Forecasting Crime 10.3 Ensembles 10.4 Bootstrap Aggregating (Bagging) 10.5 Limitations 10.6 Summary 11. Neural Networks 11.1 Building a Brain 11.2 Example: Recognizing Handwritten Digits 11.3 Components of a Neural Network 11.4 Activation Rules 11.5 Limitations 11.6 Summary 12. A/B Testing and Multi-Armed Bandits 12.1 Basics of A/B testing 12.2 Limitations of A/B testing 12.3 Epsilon-Decreasing Strategy 12.4 Example: Multi-Arm Bandits 12.5 Fun Fact: Sticking to the Winner 12.6 Limitations of an Epsilon-Decreasing Strategy 12.7 Summary Appendix A. Overview of Unsupervised Learning Algorithms B. Overview of Supervised Learning Algorithms C. List of Tuning Parameters D. More Evaluation Metrics Glossary Data Sources and References About the Authors