Island of Music from Last.FM's Playground. Visualization of listing habits based on tags
I just stumbled upon this great visualization on Last.fm Playground site. It maps listing habits based on tags as Islands. It’s not beautiful as such, but very informative and good at communicating a lot of data to viewers. Go and have a try with the interactive image.
The islands of music playground demonstration is something like a tag cloud where similar tags are located close to each together. The map was created using clustering algorithms.
These algorithms group similar music on islands. Similar islands are placed close to each other. For example, various flavours of metal are located close to each other in the upper right of the map. The map also suggests several more or less continuous transitions. For example, there is a path from folk to doom metal (via psychedelic, progressive rock, and progressive metal).
I particularly like the visualization of the paths between the different tags; “… there is a path from folk to doom metal…”. The big genres or “heavily used tags” and are placed at the edges of the map. Those genres that most people would normally use as an answer to the question, “What kind of music do you listen to?”. Musical clusters, if you like. And I might be miss- or over-interpreting here, but the smaller islands in the middle seems like commonly used bridges between these clusters (e.g. jazz, new age, soundtrack/instrumental, classical/instrumental). For-example if you listen to hip-hop and psychedelic the chances of you also listing to jazz is quite high.
There is of course a lot of noise in the data (mistagged artist, irish, comedy) like they also point out, but none the less I really like it.
