Prerequisites
Familiarity with R. Previous experience with the statnet packages (ergm, tergm, network and networkDynamic) is helpful but not required.
Description
This workshop provides an introduction to exploring temporal network data in statnet, using the R packages networkDynamic, ndtv and tsna. The software can be used for both empirical networks and simulated network data. The topics covered include:
- how to import, transform and extract relational data with timing information from various data structures (matrices, spell lists, toggles, etc)
- advantages of various temporal models and representations (continuous vs discrete time, etc)
- slicing and aggregating time in temporal networks
- calculating temporal network statistics using the tsna package, including temporal paths and basic sequence measures
- visualizing temporal networks using non-animation techniques such as relationship timelines, filmstrips and other projections
- visualizing network dynamics using animation (network movies), including exporting network animations as videos or interactive HTML5 web pages.
- attaching and manipulating dynamic vertex and edge attributes and effectively using a range of graphical properties to represent them (color, shape, size, transparency, speed, and annotation).
- common visualization challenges, such as adjustments needed when working with networks with disconnected components
- how to determine if a network has appropriate size and density to create an animation.