Nitish Singh / Yashavant Kanetkar
Librería Samer Atenea
Kálamo Books
Librería Elías (Asturias)
Librería Kolima (Madrid)
Librería Proteo (Málaga)
Data is everywhere, but data by itself has little value unless we can learn from it. Let us Learn Machine Learning takes you on a step-by-step journey from raw data to intelligent predictions. Beginning with data preparation and exploration, the book explains how to build, evaluate, and improve machine learning models using techniques such as Linear Regression, Logistic Regression, SVM, KNN, Naive Bayes, Decision Trees, Bagging, Boosting, and more. Along the way, you will learn how to engineer features, select the right models, reduce dimensionality, and deal with real-world challenges such as overfitting and imbalanced datasets. Written in a simple, practical style, this book focuses on developing intuition as much as technical skill, making machine learning accessible to students, developers, and professionals alike. Each chapter contains:● Lucid explanation of the concept.● Well thought-out, fully working programming examples.● End-of-chapter exercises to practice the skills learned in the chapter.WHAT YOU WILL LEARN● Build a complete, end-to-end ML pipeline - ingest data from CSV, SQL, APIs, and web scraping; clean and preprocess it; and explore it through univariate, bivariate, and multivariate EDA.● Engineer better features - apply encoding, feature scaling (standardization and normalization), transformations (log, square-root, Box-Cox), missing-value imputation, and outlier detection.● Master the core supervised algorithms - Linear and Logistic Regression, SVM, KNN, Naïve Bayes, and Decision Trees, each built up from intuition to math to working Python code.● Control overfitting and boost accuracy - understand the bias-variance trade-off, apply Ridge/Lasso/Elastic Net regularization, and combine models with bagging, Random Forests, and boosting (XGBoost, LightGBM, CatBoost).● Evaluate, tune, and go beyond labels - choose the right metrics (precision, recall, F1, ROC-AUC, R²), use cross-validation and hyperparameter tuning while avoiding data leakage, and uncover hidden structure with dimensionality reduction (PCA) and clustering (K-Means, hierarchical).WHO THIS BOOK IS FORThis book is for anyone beginning their machine learning journey - undergraduate and graduate students, software developers and engineers, data analysts, aspiring data scientists, and working professionals switching careers. If you know basic Python and high-school math and want to build real intuition alongside practical skills, this book is for you. No advanced mathematics required.
