Table of Contents Quick Reference

The purpose of this section is to provide a quick index of topics covered in this document so that you can easily identify and work on specific things. Use the cross-reference links below to find what you are looking for.

Welcome

Welcome

• How Should I Use this Online Companion?
• Reproducibility
• Computational Archaeology Discord Community
• New to R and R Studio?
• Contribute to the Project
• Help Build the Community
• Acknowledgements

PART II: GETTING STARTED

Section 1: Getting Started with R

Getting Started with R

• Download and Install R
  • Windows
  • MacOC
  • Linux
• Download and Install R-Studio
• Run R-Studio
• R and R-Studio Basics
  • Organization of R-Studio
  • Mathematical Operations
  • Creating Variables/Objects
  • Logical Operators
  • Vectors
  • Using Basic R Functions
  • Tabular Data
  • Data Types in R
  • Object Types in R
    • Vectors
    • Matrices
    • Data Frames
    • Lists
• The Workspace Tab
  • Setting the Working Directory
  • Working with your first R script
• Installing and Using Packages
• Working with External Files
• Plotting Data
• Warnings and Messages in R
• More Advanced R Features
  • Conditional Statements
  • Loops
  • Custom Functions
• Test Your Skills

PART II: WORK WHILE YOU READ

Section 2: Data and Workspace Setup

Data and Workspace Setup

• Data Sets
  • Just Give Me Everything
  • Roman Road Networks
  • Southwest Social Networks Project Ceramic Similarity Networks
  • Cibola Region Technological Similarity Networks
  • Himalayan Visibility Networks
  • Archaeological Publication Networks
  • Iron Age Sites in Southern Spain
• Importing Data in R
• Required/Suggested R Packages
  • Should I Just Install Everything
• R Environment
• Suggested Workspace Setup

Section 3: Network Data in R

Network Data in R

• Network Data Formats
  • Edge list
  • Adjacency list
  • Adjacency matrix
  • Incidence matrix
  • Node and edge attribute data
• Network Types
  • Simple networks
  • Directed networks
  • Signed, Categorized, and Weighted Networks
  • Two-mode networks
  • Similarity networks
    • Brainerd-Robinson similarity
    • Morisita’s Index
    • \(\chi^{2}\) Distance
    • Jaccard Similarity
    • Creating network objects from similarity matrices
  • Ego networks
  • Multilayer networks
• Converting among network object types

Section 4: Exploratory Network Analysis

Exploratory Network Analysis

• Example Network Objects
• Calculating Network Metrics in R
• Network Centrality
  • Degree Centrality
  • Betweenness Centrality
  • Eigenvector Centraoity
  • Page Rank
  • Closeness Centrality
  • Hubs and Authorities
• Triads and Clustering
  • Triads
  • Triad Census
  • Transitivity
• Walks, Paths, and Distance
  • Distance
  • Shortest Paths
  • Network Diameter
• Components and Bridges
  • Identifying Components
  • Cutpoints
  • Bridges
• Cliques and Communities
  • Cliques
  • K-Cores
  • Cluster Detection Agloritms
    • Girvan-Newman Clustering
    • Walktrap Algorithm
    • Louvain Modularity
    • Calculating Modularity of Partitions
    • Finding Edges Between Communities
• Case Study: Roman Roads in the Iberian Peninsula

Section 5: Quantifying Uncertainty

Quantifying Uncertainty

• R Scripts for Conducting Uncertainty Analysis
• A General Approach to Uncertainty
• Nodes Missing at Random
• Edges Missing at Random
• Assessing Individual Nodes/Edges Missing at Random
• Nodes/Edges Missing Due To Biased Sampling
• Resampling with Incidence Matrices
• Edge Probability Modeling
  • Edge Probability and Similarity Networks
• Uncertainty Due to Small or Variable Sample Sizes

Section 6: Network Visualization

Network Visualization

• Data and R Setup
• Visualizing Networks in R
  • network package
  • igraph package
  • ggraph package
• Network Visualization Options
  • Graph Layouts
    • Manual Layouts
    • Geographic Layouts
    • Shape-based and Algorithmic Layouts
  • Node and Edge Options
    • Nodes
    • Edges
    • Labels
  • Be Kind To the Color Blind
  • Communities and Groups
• Replicating the Book Figures
  • Figure 6.1 - Manual Layout
  • Figure 6.2 - Examples of Common Network Plot Formats
  • Figure 6.3 - Examples of Rare Network Plot Formats
  • Figure 6.4 - Simple Network with Clusters
  • Figure 6.5 - Interactive Layout
  • Figure 6.6 - Absolute Geographic Layout
  • Figure 6.7 - Distorted Geographic Layout
  • Figure 6.8 - Graph Layout Algorithms
  • Figure 6.9 - Heirarchical Graph Layouts
  • Figure 6.10 - Be Kind To the Color Blind
  • Figure 6.11 - Node Symbol and Color Schemes
  • Figure 6.12 - Image for Node, Example 1
  • Figure 6.13 - Image for Node, Example 2
  • Figure 6.14 - Edge Thickness and Color
  • Figure 6.15 - Edge Direction
  • Figure 6.16 - Edge Binarization
  • Figure 6.17 - Edge Bundling
  • Figure 6.18 - Group-in-a-box Layout
  • Figure 6.19 - Weighted Adjacency Matrix
  • Figure 6.20 - Nodetrix Diagram Interactive Plot
  • Figure 6.21 - Temporal Change, The Filmstrip Approach
  • Figure 6.22 - Temporal Change, Similtaneous Display
  • Figure 6.23 - Timelines and Time Prisms
  • Figure 6.24 - Animation
  • Figure 6.25 - Interactive Networks
  • Figure 6.26 - Case Study, SWSN Example 1
  • Figure 6.27 - Case Study, SWSN Example 2

Section 7: Spatial Networks

Spatial Networks

• Working with Geographic Data in R
• Example Data
• Planar Networks and Trees
  • Evaluating Planarity
  • Defining Trees
• Spatial Network Models
  • Relative Neighborhood Networks
  • Gabrial Graphs
  • Beta Skeletons
  • Minimum Spanning Trees
  • Delaunay Triangulation
  • K-Nearest Neighbors
  • Max Distance Networks
• Case Studies: Spatial Networks and Networks in Space
  • Proximity of Iron Age Sites in Southern Spain
  • Networks in Space in the U.S. Southwest

PART III: GOING BEYOND THE BOOK

Section 8: Exponential Random Graph Models

Exponential Random Graph Models

• ERGMs in R
• Cranborne Chase Visibility Network Example
  • Assessments of Network Properties
  • Fitting Models with ergm
  • Building a Model Based on Theory
  • Assessing Goodness-of-Fit
  • Asessing Models and MCMC Diagnostics
• Simulating Networks from ERGMs
• Additinoal Info on ERGM Terms
  • Avoiding Model Degeneracy

Section 9: Spatial Interaction Models

Spatial Interaction Models

• Simple Gravity Models
  • Parameterizing the Gravity Model
• The Rihll and Wilson “Retail” Model
  • Parameterizing the Retail Model
• Truncated Power Functions
• Other Spatial Interaction Models

Section 10: Affiliation Data and Co-Association

Affiliation Data and Co-Association

• Analyzing Two-Mode Networks
  • Using Traditional Network Metrics
  • Using Two-Mode Specific Network Metrics
  • Projecting Two-Mode Networks Before Analysis
• Correspondence Analysis
  • Network Visuals Using Correspondence Analysis
• Measuring Co-Association
  • Alternative Methods for Visualizing Co-associations

Section 11: Network Diffusion

Network Diffusion

• Diffusion Processes
• Simulating Network Diffusion in R
  • Simulated Networks
  • Empirical Networks
• Evaluating Diffusion Models

Section 12: Comparing Networks

Comparing Networks

• Known Node Correspondence
  • Direct Comparison of Adjacency Matrices
  • Quadratic Assignment Procedure
  • DeltaCon
• Unknown Node Correspondence
  • Comparing Network Global and Local Statistics
  • Graph Kernel Methods
  • Spectral Methods
  • Portrait Divergence
  • Graphlet-based Methods
  • Alignment-based Methods

ADDITIONAL INFORMATION AND MATERIALS

References and R Packages

ArchNetSci GitHub Repository