User Tools

Site Tools


Visual Analytics Course

Bordeaux roadmap 2012-2013 / Georges Nader's Homework 1

Critères & points attribués Points selon votre argumentation (aspects positifs/négatifs)
(1) Présentation / Objet et objectifs 1
(2) Distance opérationnelle 1,5
(2) Distance articulatoire 1,5
(2) Distance sémantique 1,5
(7) Total 5,5

G.M. Les notes attribuées tiennent compte de la prestation orale en cours.

Guy Melançon 2012/10/14 18:56

I chose to talk about the "Scientific Collaboration" map computed by Olivier H. Beauchesne. This map represents the scientific collaboration between cities around the world. The data was taken from published articles in various journals and databases and was build using the following rule : If a researcher from UCLA published a paper with someone from NYU a collaboration between New York and Los Angeles is added.

G.M. Well done, your short intro gives us all the necessary details to grasp the designer's intention.

Using the geographic location of a city as its representation reduced the articulatory distance of the map. Even with the high resolution version that does not fit on the screen it is very easy to find a city because we are all familiar with the geographic map of the world. G.M. I partly agree; this has to do with the semantics of the map. In addition this geographic representation improved the map's readability and using color intensity to repesent the number made it even more easy G.M. Agree. This takes place at the operational level (pre-attentive process).. You don't have to put any effort to extract important piece of information like extracting the most important scientific poles in the world, a region, a country; In France Paris is the hub of most research. G.M. One could argue that this is where semantics helps reducing articulatory distance. However, I did find a very important ambigus element in this map that made me question and this about its meaning. The color intensity between two cities is a function of the logarithm of the number of collaborations betweena pair of cities and the logarithm of the distance between those same two cities. This means if two cities A and B have the same number of collaboration with a third one C but A is closer to C, it would appear that A made more collaboration with C than B made with C. At first I was analysing the map by saying that in the German/Netherlan area the most research is being done so the most important Universities must be there, but then I realized the reason that this area a brighter is that maybe because there are too much small universities there and they collaborate too often with each other; I can't be sure of both hypothesis. G.M. So you're saying the is a //semantic gap// (semantic distance gets higher).

G.M. I would have liked you to adopt a more structured comment based on the three levels of 'gaps'. See the general comment on Homework 1.

/net/html/perso/melancon/Visual_Analytics_Course/data/pages/gnader/solution_homework1.txt · Last modified: 2012/10/14 18:56 by melancon