Spis treści: :: Library Catalog

Zapisane w:

Opis bibliograficzny
Główni autorzy:	DR. KOUSHIK MANDAL, SONALI PARUA
Format:	Recurso digital
Język:
Wydane:	Zenodo 2025
Dostęp online:	https://doi.org/10.5281/zenodo.15879471
Etykiety:	Dodaj etykietę Nie ma etykietki, Dołącz pierwszą etykiete!

Spis treści:

In recent years, we have witnessed a profound transformation  in the way data is generated, stored, analyzed, and ultimately utilized  to inform decisions across nearly every domain of human activity.  This era of Big Data has catalyzed an unprecedented synergy between  traditional statistical methods and the fast-evolving field of machine  learning. As organizations and researchers navigate massive  datasets—from consumer behavior logs to genomic sequences—the  ability to extract meaningful insights depends increasingly on a robust  statistical foundation interwoven with computational efficiency and  algorithmic sophistication.  This book, Statistical Methods for Big Data and Machine  Learning, is designed to bridge that critical intersection. It provides an  integrated view of the statistical underpinnings essential for modern  data science, while also offering practical insights into how these  concepts manifest within machine learning pipelines, tools, and real world applications. Whether applied to model credit risk, optimize  marketing strategies, understand public health trends, or power  recommendation systems, the statistical methods covered here are  fundamental to extracting value from data at scale.  The early chapters establish a foundational understanding of  statistics—both descriptive and inferential—before progressively  moving toward advanced topics such as regularization, resampling,  Bayesian inference, and ensemble learning. The book addresses  essential questions: How do we model uncertainty in large-scale  systems? How do we select features from high-dimensional data?  What are the ethical considerations when deploying models in  sensitive contexts? Each chapter not only introduces theoretical  frameworks but also emphasizes implementation, interpretability, and  critical thinking, ensuring readers are not only capable of using  statistical methods, but also of questioning their limitations and  implications. The data analysis section delves into the intricacies of  working with structured, semi-structured, and unstructured data.  Concepts such as estimates of location and variability, as well as  categorical data analysis, are covered in depth. A dedicated segment  on database systems, including relational databases, NoSQL, and  query optimization, ensures that readers understand the infrastructure  behind data storage and retrieval.  A key strength of this text lies in its holistic structure. We  have organized the material to suit a wide audience: undergraduate and  graduate students in statistics, data science, or computer science;  professionals in industry seeking to deepen their analytical skills; and  educators looking for comprehensive teaching material. From a  thorough discussion of machine learning’s statistical roots to hands on case studies rooted in global and Indian contexts, this book aims to  be as accessible as it is rigorous.  The journey from raw data to actionable insight is not merely  a technical one; it is also conceptual, ethical, and deeply human. As  data volumes continue to grow, so too must our responsibility to use  them wisely. Our hope is that this book serves not just as a guide to  methods, but also as a catalyst for responsible, impactful data-driven  discovery.

Podobne zapisy