Applied Machine Learning with Python

If you are looking for an engaging book, rich in learning features, which will guide you through the field of Machine Learning, this is it. This book is a modern, concise guide of the topic. It focuses on current ensemble and boosting methods, highlighting contemporray techniques such as XGBoost (2016), Shap (2017) and CatBoost (2018), which are considered novel and cutting edge models for dealing with supervised learning methods. The author goes beyond the simple bag-of-words schema in Natural Language Processing, and describes the modern embedding framework, starting from the Word2Vec, in details. Finally the volume is uniquely identified by the book-specific software egeaML, which is a good companion to implement the proposed Machine Learning methodologies in Python. APPLIED MACHINE LEARNING WITH PYTHON 1 Contents 6 List of Figures 10 Preface 14 Chapter 1. Introduction to Machine Learning 20 1.1 A simple supervised model: Nearest Neighbor 20 1.2 Preprocessing 33 1.3 Methods for Dealing with Imbalanced Data 44 1.4 Reducing Dimensionality: Principal Component Analysis 51 Chapter 2. Linear Models for Machine Learning 60 2.1 Linear Regression 60 2.2 Shrinkage Methods 62 2.3 Robust Regression 70 2.4 Logistic Regression 76 2.5 Linear Support Vector Machine 86 2.6 Beyond Linearity: Kernelized Models 92 Chapter 3. Beyond Linearity: Ensemble Methods for Machine Learning 102 3.1 Introduction 102 3.2 Ensemble Methods 102 3.3 Random Forests 110 3.4 Boosting Methods 114 Chapter 4. An Introduction to Modern Machine Learning Techniques 134 4.1 Introduction to Natural language Processing 134 4.2 Introduction to Deep Learning 158 Appendices 170 Appendix A. A crash course in Python 172 A.1 Building Blocks in Python 172 A.2 Data Structure in Python 174 A.3 Loops in Python 178 A.4 Advanced Data Structure in Python 180 A.5 Advanced Concepts on Functions 182 A.6 Introduction to Object-Oriented Programming 188 Appendix B. Mathematics behind the skip-gram model 194 Index 196 Bibliography 198 Back Cover 204