Turn on ye ole Javascript to add ratings in this low-budg app.

Title:

Help! My iPod Thinks I'm Emo

Your vote:
Yes No
Organizer:
Paul Lamere, Sun Labs, Sun Microsystems Inc.
Description:
Music recommendation is broken - automatic music recommenders make mistakes that no human would ever make. In this talk, we will explore why recommenders make such dumb mistakes and we will explore some of the new ideas coming from recommendation and music researchers to help make music recommendations better.
Questions
Answered:
  1. Why is music recommendation important?
  2. How do current recommenders like iTunes Genius, Last.fm and Pandora work?
  3. What are the strengths and weaknesses of collaborative filtering?
  4. How can content-based recommendation techniques be used to extend recommendations into the long tail?
  5. How can music recommenders be evaluated?
  6. How can a recommender be immune to hackers and shillers?
  7. What are some of the new ideas coming from the academic world that will help make recommenders better?
  8. How can user interface and visualizations improve the music discovery experience?
  9. Why are transparent recommendations (recommendation with an explanation) so important?
  10. Why are novelty and context so important to music recommendation?
Panelists:
Paul Lamere (Sun Microsystems), Anthony Volodkin (The Hype Machine)
Level:
Advanced
Category:
New Technology / Next Generation
Type:
Solo
Event:
SXSW Interactive 2009
james miao
on 6/10/08
Obviously a pressing issue, especially considering how people (subjective creatures!) usually demonstrate low-tolerance for "bad" recommendations.
on 18/5/09
Interesting, you have my vote.
on 23/7/09
Not an easy thing to sort out but who knows why people recommend things.
It certainly is difficult to evaluate music recommenders and their services. They play such a huge part is whats hot musically and moving forward this will only get more so as we move forward.
on 11/10/09
Music recommendation is important because good things should be known to all.
on 24/10/09
Not an easy thing to sort out but who knows why people recommend things.
on 8/11/09
Automatic music recommendations by machines can never be compared to the accuracy of humans.
on 13/11/09
Sounds like it will be a very interactive panel... like the style.
on 26/11/09
very Interesting discussion, Thank you
Developed for SXSW by Lindsey Simon