Data Science Under Uncertainty

Inference, Prediction, and Decision-Making with Python
Erscheinungsjahr 2027
ISBN: 978-3-032-43013-7
Verlag: Springer

Buch, Englisch, 326 Seiten, Format (B × H): 155 mm x 235 mm

This book provides a practical introduction to data science as decision-making under uncertainty. It shows how to move beyond point estimates and accuracy scores toward workflows that quantify, evaluate, and communicate uncertainty in real data problems. Readers learn probability as a language for data, simulation-based thinking, regression and predictive modeling, resampling and regularization, probabilistic classification, calibration, prediction intervals, Bayesian modeling, conformal prediction, causal questions, decision analysis, robustness under distribution shift, and responsible communication. With Python-oriented examples and end-to-end case studies, this book helps readers build models whose limitations are explicit and whose predictions can support better decisions.

Zielgruppe

Professional/practitioner

Autoren/Hrsg.

Unpingco, José

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Maschinelles Lernen
• Mathematik | Informatik Mathematik Stochastik
• Mathematik | Informatik EDV | Informatik Daten / Datenbanken

Inhaltsverzeichnis

Introduction.- Part I: Foundations for uncertainty-aware data science.- Data science is decision-making under uncertainty.- Probability as a language for data.- Part II: Estimation, modeling, and validation.- Estimation, sampling, and simulation.- Regression and predictive modeling.- Uncertainty in model fitting.- Resampling, cross-validation, and regularization.- Part III: Probabilistic prediction and uncertainty quantification.- Probabilistic classification.- Calibration and prediction intervals.- Bayesian modeling for data science.- Conformal prediction.- Part IV: Causality, decisions, deployment, and communication.- Causal questions in data science.- Decisions, losses, and utility.- Robustness and distribution shift.- Communicating results under uncertainty.- Projects and case studies.- Conclusion.

Über Autor(innen)

Dr. José Unpingco completed his PhD from the University of California, San Diego (UCSD) and has since worked in industry as an engineer, consultant, and instructor on a wide-variety of advanced data science topics, with deep experience in machine learning. He was the onsite technical director for large-scale Signal and Image Processing for the Department of Defense (DoD) where he also spearheaded the DoD-wide adoption of scientific Python. In his time as the primary scientific Python instructor for the DoD, he taught over 600 scientists and engineers. Dr. Unpingco is currently a Senior AI Engineer at ServiceNow and an award-winning senior lecturer at UCSD, teaching in the undergraduate and graduate Machine Learning and Data Science degree programs.