Here is a first movie I made on neuromarketing.
To explain a bit: this is realized with Gephi, a software for the visualization of networks evolving through time.
What is the data represented? This is a co-occurrence network, obtained through the AlchemyAPI plugin for Gephi. I know, this sounds cryptic! Basically, the idea is that words represented in the network are the most frequent ones used in webpages discussing neuromarketing. A link between any two nodes means that those two words frequently co-occur (ie, they are used in same documents, as opposed to being frequently used but never appearing simultaneously in the same text).
This is just a proof of concept – the movie is hard to read and needs much refinement. I presented today at the Sunbelt conference a refined version, with accompanying slides (do click on the slides, not on the timeline below them).
From my lay vantage point, I would need a marker telling how fast time is passing, and also some markers telling me what to look at. I mean, of the kind I found in your slides (green, blue circles and archs, etc.). It would help me identify how such and such community or subject is specifically evolving, or how one community is dissolving while another is expanding, ect.
But nice work already.
Thanks!
You are right. Even more than for static visuals, the data does not speak by itself when presented in movies (naively, I expected the contrary). The viewer needs a “curator” or a narrator.
I plan to do this the following way: with Adobe Captivate, which can be also be used for video editing, especially for “software demos” or “online learning animations”. It has all the functions to add arrows, circles, buttons, titles and legends, and other bubbles, to an existing video. The downside of it is that we are talking video editing: takes a lot of time. I’ll do that when I’ll be ready to submit.
hi clement,
looks very interesting. would love to discuss the methods with you, once I actually started with the Slow Food project (this May). I had never heard of this Gephi-softwared, but it looks very interesting.
best,
stefan
Hi Stefan,
With pleasure. Do contact me when you are ready, we’ll see what kind of data you have, and how a visualization could help explore your research questions!