By performing a temporal selection at this new resolution, we begin to see what happens at the start of the outbreak. To identify ground zero we can change the resolution of the bins to a much finer-grained analysis. The timeline shows data grouped into bins of 6 hours.
Visualization mini lesson how to#
We can do a temporal selection on the data to find out how to disease begins to spread from that point. The timeline shows very clearly when the epidemic first starts, about the 18th of May. This led us to filter our data so that only the first entry from each poster was shown on the graph, here shown by red bars, and on the map we see that there are no longer any concentrations around the main hospitals, indicating that people first posted when they became ill, away from the hospitals. Here we see all posts by the same poster, indicating that they’d tweeted several times about the same illness. We could confirm this by examining the history of the people who tweeted on the map. The distribution of posts shown on the map indicate a concentration around the hospitals, this led us to believe that at least some of these posts were second or third entries from people who’d already fallen ill elsewhere. Our main application comprised three views of the blog posts, firstly one showing where they occurred, secondly one showing when they occurred, including the associated weather over this timeline, and thirdly, the posts themselves. The first stage of our analysis involved identifying the key words and phrases that we thought were associated with the epidemic, this allowed us to select only those blog entries that we thought were relevant for the analysis of the disease.
They also focus on revealing unknowns (i.e., generating insights), rather than communicating known trends. Visual analytic tools typically belong in the opposite corner-these tools are characterized by high human-map interaction and are often designed with private data or data that is otherwise meant for domain experts. Most of the maps we have designed thus far would be considered to be in the communication (public, static, and intended to present known information) corner of the cube.
Recall the Cartography Cube from Lesson 1 (review this concept in the Communicating with Maps section). Interactive mapping has played an important role in the field of visual analytics, defined as “the science of analytical reasoning facilitated by interactive visual interfaces” (Thomas and Cook 2005). When we talk about interactivity in maps, we must consider not just user interactivity within maps, but interactively among maps, as well as with other tools and visual graphics.
0 kommentar(er)
