User login

Joel Oughton's blog

28

May

2011

Had a play around with the google maps api. I need to visualise exports from a given country (starting with NZ) to all other countries for a group project. I turned a list of countries in to longitudes and latitudes using the google geolocation api. Was rather easy - just loop through all of the countries and make an api request in a rate limited way. Then, using export data from stats New Zealand, I drew weighted lines from NZ to all other countries. The resulting shape kind of looks like a butterfly.

View image

This is only the first iteration. The next step is to draw paths through the ocean that split off into countries. Similar idea to Minard's famous French wine exports drawing:

View Image

I also made a start on my 520 interim report, polished off two assignments and made a couple improvements to my network map.

20

May

2011

I made it easier to move between nodes in my network map, while zoomed in. You can now move along edges to the destination node. I also implemented some more edge types such as double arrows to indicate when edges are connected in both directions.
Read about 5 papers on efficiency related to topology inference from comp513. It was interesting to link theory I have learned in my data mining paper (information theory) to traceroute efficiency. Still have 3 more sections to write for that assignment (Alias resolution, ISP mapping and a section of my choosing).
Planned out a visualisation for my info vis paper that will display the exports of a given country with respect to the rest of the world.

13

May

2011

This weeks focus was on making a start on 3 major assignments (30% +) that are due by the end of the semester.

I have a few more ideas to work on with my 520 project. I met with Jaime and he gave me a good overview of his core network at layers 2 and 3. Still cloudy as to what a good way of bringing multiple layers together in a single visualisation will look like but it is good to have more examples to work with.This weeks focus was on making a start on 3 major assignments (30% +) that are due by the end of the semester.

I have a few more ideas to work on with my 520 project. I met with Jaime and he gave me a good overview of his core network at layers 2 and 3. Still cloudy as to what a good way of bringing multiple layers together in a single visualisation will look like but it is good to have more examples to work with.

Started mapping out a bunch (~20) of papers related to internet topology discovery for the 513 paper. Seemed to pull a good list of relevant papers together to read for my lit review.

Implemented an random forest type machine learning algorithm in java (called Extra Tree - http://www.montefiore.ulg.ac.be/~ernst/extremely-randomized-trees.pdf) for another assignment. I have regression mostly working, need to tweak it and work for classifications too.

Started mapping out a bunch (~20) of papers related to internet topology discovery for the 513 paper. Seemed to pull a good list of relevant papers together to read.

Implemented an random forest type machine learning algorithm in java (called Extra Tree) for another assignment. I have regression mostly working, need to tweak it and work for classifications too.

06

May

2011

This week (and some of last week) I took the Karen weathermap and implemented it using my current network map visualisation. I used a static layout for the POPs and just a basic star layout for regional devices connected to these POPs. The layouts can be applied to a given subnetwork's nodes and can be changed at runtime. As you zoom in on a POP, nodes and links connected become more visible and labels appear. The underlying concept being, that subnetworks can contain subnetworks that contain subnetworks etc.