 |
|
|
 |
|
|
|
|
Graphs
Visualizations From A Different Perspective
When we want to create a visualization that shows relationships involving a group of things, then there is a special family of visualizations at our disposal: "graphs".
What is a graph?
A graph is a specific type of visualization, often used to show direct relations between objects in a group. These objects can be most anything: people, or documents, or softball teams, or really most any thing that can have a relation to another thing. In a graph visualization, each of these objects is usually depicted as a point (or, in mathematical lingo, a node or vertex – plural vertices).
Graphs vs. Charts
A Chart
|
"Graph" is a mathematical term and we use it in a more general way than that which you probably remember from school and likely often see being used to show information like market or polling data in newspapers and magazines. To keep these two things distinct, we use the word chart to describe this kind of illustration.
Why should you care about graphs?
Just as visualizations, in general, present certain types of information much more clearly than text and printed numbers, graph visualizations present certain types of information much more clearly than any other types of visualizations. Graphs are especially useful in helping human vision in recognizing patterns of grouping (or 'clustering') in large populations of inter-related objects.
As a thought experiment, let's assume that you have all of the emails sent and received within a hypothetical company of 250 employees during one month. It is easy, but nonetheless revealing, to create a graph with each person represented by a vertex; should the person have exchanged emails with another person in the company, then there would be a line (or edge) joining the two vertices representing those people. One could add visual cues by making the line thicker, or a different color, or drawing their vertices on top of other vertices, to highlight important relationships such as people who participated in a certain discussion, or were part of a relevant workflow, or who communicated on particular topics.
You can see how this would allow a person to quickly glance at the graph visualization and nearly instantly grasp the story of the email social network within this hypothetical company.
For even more revealing, slightly less easy, visualization, such a graph could be brought to life by examining the growth and changes as time progressed through the life span of the company for which you possess the emails; you can imagine each moment in time being like a frame in a movie, and the movie allowing you to see the thriving and the atrophy of communication over time.
Imagine including in this movie a timeline which is synched to known public events involving the company - now we're able to depict an even more interesting recreation of historical events within this hypothetical company.
The visual arts and sciences
There is more than a little bit of art accompanying the science of making a good visualization and this is where one of Cataphora's strong suits comes into play for you.
After we've performed the analysis on your data in order to gather the information that gives birth to the graph visualization, the first step of basic visual positioning must take place. Laying out the graph - deciding where each vertex and edge should be positioned on the screen in order to most effectively convey the information - is an extremely complex process. It requires an immense amount of computer processing power, and is the subject of continual academic and industry research; indeed, there are software companies that do little more than develop and sell libraries to do this work.
Most graph layouts algorithms focus on laying out the graph solely on the basis of fundamental attributes of the vertices and edges. By contrast, Cataphora employs algorithms and heuristics (rules of thumb) that strive to produce even more helpful layouts by including algorithmic hints that describe the noteworthy information contained within the graph. These added hints allow our computers to produce graphs that convey the message more starkly than would be possible were with conventional algorithms.
The Cataphora difference
Cataphora's visual engineers are well versed in creating the graphs with the greatest amount of impact. Our team of engineers possesses the mathematical know-how to choose and roughly shape the marble that will be your graph visualizations, as well as possessing the artistic dexterity to perform the finishing touches. You can look forward to Cataphora delivering visualizations of your data in clean and clear manners. To help you get there, your Cataphora client services team will work with both you and our engineering teams to bring out what you need.
|
|
|
