Data Science Project: An Inductive Learning Approach
Filipe Alves Neto Verri First Edition — v1.0.0 “Skyward Vector” — February 2026
About the Book
This book provides a structured exploration of the entire data science pipeline, from data collection to model deployment. It effectively balances theory and practice, focusing on the inductive principles underpinning predictive analytics and machine learning. Unlike texts that center on algorithms or specific tools, this book provides a holistic, language-agnostic view of every stage of a data science project.
Born from lecture notes for the graduate course PO-235 Data Science Project at ITA (Aeronautics Institute of Technology) and UNIFESP (Federal University of São Paulo), the book serves as a textbook for data science project courses and as a reference for professionals. It does not teach specific algorithms; instead, it explains why machine learning works, increasing awareness of its pitfalls and limitations.
The scope is deliberately focused on predictive and inductive methods, with deep attention to correct evaluation and validation of data science solutions. A solid mathematical and statistical foundation is expected from the reader.
Table of Contents
| # | Chapter | Description |
|---|---|---|
| 1 | A Brief History of Data Science | Origins and evolution of the field |
| 2 | Fundamental Concepts | Core definitions and theoretical foundations |
| 3 | Data Science Project | Structure and lifecycle of a DS project |
| 4 | Structured Data | Representations, types, and semantics of data |
| 5 | Data Handling | Collection, storage, and data quality |
| 6 | Learning from Data | Inductive learning principles and theory |
| 7 | Data Preprocessing | Transformation, normalization, and feature engineering |
| 8 | Solution Validation | Evaluation protocols, metrics, and statistical tests |
| A | Mathematical Foundations | Appendix — Linear algebra, probability, and statistics |
For Instructors: Companion Slides
Companion slide decks are available for classroom use. The slides follow the book’s structure and reproduce all original TikZ figures.
License: The slides are released under CC BY-NC 4.0 — you may adapt them for teaching purposes (more permissive than the book’s CC BY-NC-ND 4.0 license).
| # | Slide deck | Download |
|---|---|---|
| 1 | A Brief History of Data Science | |
| 2 | Fundamental Concepts | |
| 3 | Data Science Project | |
| 4 | Structured Data | |
| 5 | Data Handling | |
| 6 | Data Exploration | |
| 7 | Learning from Data | |
| 8 | Data Preprocessing | |
| 9 | Solution Validation | |
| A | Mathematical Foundations |
LaTeX source for the slides is available on GitHub.
Citation
@book{verri2026datascienceproject,
author = {Verri, Filipe Alves Neto},
title = {Data Science Project: An Inductive Learning Approach},
year = 2026,
publisher = {Leanpub},
address = {Victoria, British Columbia, Canada},
doi = {10.5281/zenodo.14498010},
url = {https://leanpub.com/dsp},
note = {Version v1.0.0}
}