Thinking Data Science: A Data Science Practitioner's Guide

This book provides a comprehensive guide to Machine Learning projects, covering technology selection, model development, and handling large datasets. It offers a systematic approach to problem-solving and consolidates available algorithms and techniques for efficient model design.

Format: Hardback
Length: 358 pages
Publication date: 02 March 2023
Publisher: Springer International Publishing AG

This comprehensive guide to Machine Learning projects addresses the challenges faced by aspiring or experienced data scientists, including:

Confusion about the appropriate technology for ML development: Should I use GOFAI, ANN/DNN, or Transfer Learning?
Reliance on AutoML for model development: Can I trust AutoML to handle model development?
Handling large datasets: What if the client provides me with Gig and Terabytes of data for developing analytic models?
Dealing with high-frequency dynamic datasets: How do I handle datasets with frequent updates?

The book aims to provide a consolidated "cheat sheet" for the entire data science process, encompassing machine learning algorithms and neural networks. The challenge for data scientists is to extract meaningful information from vast datasets to create better strategies for businesses. Machine Learning algorithms and Neural Networks are designed to analyze such datasets.

Making a decision on which algorithm to use for a specific dataset can be daunting for data scientists. However, a systematic approach to problem-solving is necessary. This book describes various ML algorithms conceptually and discusses a process for selecting ML/DL models. The key aspect of this book is the consolidation of available algorithms and techniques for designing efficient ML models, regardless of the size of the data.

Thinking Data Science is a valuable resource for practicing data scientists, academics, researchers, and students who aim to build ML models using the appropriate algorithms and architectures. Whether the data is small or large, this book will help practitioners navigate the complexities of machine learning and develop effective solutions for real-world problems.

Weight: 716g
Dimension: 160 x 243 x 30 (mm)
ISBN-13: 9783031023620
Edition number: 1st ed. 2023