Every network graph is a recording of a dynamic process over time. No matter what the nodes and the edges represent — interactions, co-occurrence, affinity — the relations unfold over time and the graph shows the patterns that emerge during this evolution.
It can be interesting, therefore, to see not only the final stage when the graph is complete but to also observe how the graph evolves over time.
Using InfraNodus you can watch evolution of any graph over time. You can use this view to analyze how the network's structure changed over time. You can also use it to demonstrate dynamic properties of graphs in classrooms and to learn more about how graphs evolve.
There are several Dynamic View options available.
Step 1: Activate the Dynamic View option by pressing the Timer button in the Statements menu:
Activating this option will force InfraNodus to highlight the statements that you are currently scrolling through on the graph. It will also automatically recalculate all the graph statistics for it. So you will only see the part of the graph that represents the statement that you can see in the statements panel.
Step 2: Optionally, you can also launch the player, using the Timer Play button in the Statements menu. The statements will then scroll for you and the graph will "Play" out its dynamic evolution (as if you were scrolling the statements manually)
You can set the speed of the player and also choose to only highlight one statement at a time (by default 2-3 statements are highlighted (the ones that are currently visible in the Statements panel).
Step 3: You can also choose between the different dynamic visualization modes. It is possible to visualize the gradual build-up towards the final graph, to highlight the current statements in the final graph, or to highlight only the visible part, hiding the rest of the graph. Here are the different options explained below:
Default Option #1: Evolution Towards the Final Graph
Starts from the initial state, adding a graph representation of every consecutive statement, gradually building the graph to its final shape. Showing how the discourse was sculpted in time. Recalculating the graph statistics and measures at every iteration. Highlighting with blue currently visible statements, so that we see the path for the spread of ideas.
Without selecting the currently visible statements:
Note that the examples above also demonstrate that you can send an external link to your graph that will automatically launch the dynamic graph player.
Option #2: Visualize Build-Up of the Final Graph
Same as above, but always showing the whole graph, emphasizing what has already been created, and highlighting the current (most recent) statements visible. Shows the spread of ideas through the final structure.
Option #3: Visible Statements in the Graph
Starts from the initial state, showing the graph only for a certain (visible) span of the statements. Allows to see the evolution of a discourse in time, what were the topics that emerged and receded, what was the dynamics of meaning circulation (i.e. only a handful of main nodes in the center or also exploring periphery).
Option #4: Visible Time Span in the Graph
This is similar to the option above, however, instead of just showing the current statement InfraNodus will show a certain timespan going back several statements from the one that you currently see. It emulates the way we read, so you can see how the meaning evolved over time taking short-term memory into account. This is perhaps the closest representation of how we actually perceive text when we read.
Option #5: Highlight the Statements in the current Graph
If all the options above can be a bit heavy on your browser (because the graph analytics is automatically recalculated every time), you can use the simple Highlight the Statements feature which simply highlights the currently visible statement in the final graph without recalculating the graph stats.
Note, that you can embed any of those visualizations as an iframe or using a public URL to any website. Let us know about your examples of using the dynamic visualization feature!