# Working with Network Data

## Authors

- James Bagrow, University of Vermont
- Yong-Yeol Ahn, Indiana University, Bloomington

## Why this book?

This book focuses on the *practical side* of network science — **working with
network data** — to provide a more practical guide for data scientists to use network science.

We hope that this book can help researchers in day-to-day tasks, starting from the very act of conceptualizing networks through to sophisticated network analysis, from exploratory analysis to statistical modeling and machine learning. At the same time, we also aim to give data scientists a foundational understanding of the tools, both mathematical and computational, at their disposal. The breadth and depth of statistical methods we can now use on network data is dizzying. We wish to take the prepared data scientist from their base knowledge of mathematics and statistics forward on a journey through the fundamentals of network data.

## Table of Contents

## Click here to see the Table of Contents

**I. Background**

- A whirlwind tour of network science
- Network data across fields
- Data ethics
- Primer

**II. Applications, tools, and tasks**

- The life cycle of a network study
- Gathering data
- Extracting networks from data — the “upstream task”
- Implementation: storing and manipulating network data
- Incorporating node and edge attributes
- Awful errors and how to amend them
- Explore and explain: statistics for network data
- Understanding network structure and organization
- Visualizing networks
- Summarizing and comparing networks
- Dynamics and dynamic networks
- Machine learning

**Interlude — good practices for scientific computing**

- Research record-keeping
- Data provenance
- Reproducible and reliable code
- Helpful tools

**III. Fundamentals**

- Networks demand network thinking: the friendship paradox
- Network models
- Statistical models and inference
- Uncertainty quantification and error analysis
- Ghost in the matrix: spectral methods for networks
- Embedding and machine learning
- Big data and scalability