the science of Data: A precis about application data

The Science of DataA Complete Guide to AI, Machine Learning, and Data ScienceWhat does it mean to learn from data?The Science of Data is a rigorous yet accessible single-volume companion to modern data science, artificial intelligence, and machine learning. Written by a practitioner and educator with over 25 years at the forefront of AI in Silicon Valley, this book connects the mathematical foundations of intelligent systems to the concrete mechanics of building, evaluating, and deploying predictive models.The book opens with the question Alan Turing posed in 1950 - can machines think? - and systematically builds the framework needed to answer it. From the four laws of probability and Bayes' theorem to deep neural networks and transformer architectures, every topic is grounded in plain-language motivation before formal treatment.Core chapters cover: AI and Machine Learning fundamentals (intelligent agents, rationality, generative vs. discriminative methods, maximum likelihood, entropy); Foundations (probabilistic reasoning, stochastic processes, search, optimization, regularization); Supervised Learning (regression, decision trees, random forests, SVMs, neural networks); Unsupervised Learning (k-means, PCA, LDA, word embeddings); Reinforcement Learning (MDPs, Q-learning, policy gradient, genetic algorithms); Deep Learning (CNNs, transformers, transfer learning, LSTM, explainable AI); Bayesian Inference (priors, posteriors, MCMC, Metropolis-Hastings, Gibbs sampling, Stan); Support Vector Machines (VC dimension, statistical learning theory, kernel trick); The Science of Data (cognition, data-knowledge hierarchy, information theory); Knowledge and Reasoning; The Data Science Process (CRISP-DM, feature engineering, model evaluation, lift metric); and a curated A-Z glossary of 268 essential terms.Five technical appendices cover Python (NumPy, Pandas, scikit-learn, PyTorch), R (tidyverse, ggplot2, tidymodels, RStan), MATLAB/Octave/Scilab, Optimization methods, and curated Public Data Sources.Whether you are a student seeking theoretical grounding, an engineer building a unified mental model, or a researcher navigating probabilistic methods, The Science of Data provides the complete map - from the why, through the what, to the how of intelligent systems.

Pagina's: 338, Paperback, Independently published