Mathematics is the foundation of data science techniques. With the democratization of data science, almost anyone has access to easy-to-use tools and platforms to get started with data science applications. However, a serious professional or researcher would sooner or later need to understand the mathematics behind the wide variety of data science methods. To this end, this book provides a lightweight mathematics background for common machine learning models and techniques.
The book starts with a refresher in linear algebra and a concise overview of basic mathematics terms and operations, such as vectors, matrices, and eigenvalues and eigenvectors. However, the real meat of the book starts with chapter 2, where the authors introduce principal component analysis (PCA). In a later chapter, the authors cover other well-known and commonly used techniques such as k-means, classification algorithms, and tree-based classifiers. The authors first provide a brief description of the technique, to establish the need for it, and then delve into the mathematics of the concept. The authors typically include brief examples to show the transformations and applicable operations; these examples are welcome additions to the book. The provided figures and tables are clean; in general, they help readers get a better understanding of the explained concept.
The book assumes a basic understanding of mathematics notations and operations, such as summation, and does a good job of raising the bar on the mathematics knowledge associated with machine learning models. However, the authors fail to include a much-desired piece: a high-level logical explanation of data science operations without using any Greek symbols.