Archive for category tags

Using Visualizations for Music Discovery

As Ben pointed out last week, the ISMIR  site has posted the tutorial schedule for ISMIR 2009.  I’m happy to see that the tutorial that Justin Donaldson and I proposed was accepted.  Our tutorial is called Using Visualizations for Music Discovery.  Here’s the abstract:

As the world of online music grows, tools for helping people find new and interesting music in these extremely large collections become increasingly important.  In this tutorial we look at one such tool that can be used to help people explore large music collections: data visualization.  We survey the state-of-the-art in visualization for music discovery in commercial and research systems. Using numerous examples, we explore different algorithms and techniques that can be used to visualize large and complex music spaces, focusing on the advantages and the disadvantages of the various techniques.   We investigate user factors that affect the usefulness of a visualization and we suggest possible areas of exploration for future research.

I’m excited about this tutorial – mainly because I get to work with Justin on it. He’s a really smart guy who really knows the state-of-the-art in visualizations. I’ll just be tagging along for the ride.

Detail from Genealog of Pop/Rock Music

Detail from Genealogy of Pop/Rock Music

We  are in the survey phase of talk preparation now. We’ve been gathing info on various types of visualizations and tagging them with the delicious tag MusicVizIsmir2009.  Feel free to tag along (pun intended) with us and tag items that you encounter that you feel may be particularly interesting, unique or salient.



Making boring homework fun

JFs Pop Artist slide

JF's Pop Artist slide

One of the cool things about working at Sun Labs is all of the very smart interns that come and work for a summer. They are invariably creative, and bring  many new ideas to the labs.   One intern I worked with a few years back, Jean-Francois, just posted a blog entry about how he wanted to create a slide for a presentation that showed the pop artists from Sweden as a word cloud.  Now most of us would have just typed in a few artist names and resized the fonts based on what we though was the approximate popularity of the particular artist.  But not Jean Francois.  He turned this slide into its own little research project – first trying to scour the popularity data from the Wikipedia, then mining Google search results and finally settling on using the webservices to get the listener data.

JF took a rather boring assignment – make an oral presentation on Sweden – and made it a learning experience for something that was interesting to him.  I am not sure how much he learned about Sweden, but he certainly learned something about web mining, artist name resolution and ambiguity,  and  using web services.  I wonder if Jean Francois’s International Communication’s professor understood the depths of detail that JF went to get the information on that one slide correct.

Anyway, it is a cool post from a JF.

Leave a comment