Communications of the ACM
Finding Good Music in Noisy Online Markets
Credit: Christine Daniloff/MIT
Columbia University researchers began an online social-media marketing experiment in 2004, creating nine versions of a music download site that presented the same group of unknown songs in different ways. The researchers wanted to measure the effect of early peer recommendations on the songs' success, but found that different songs became hits on the different sites and that the variation was unpredictable.
Massachusetts Institute of Technology (MIT) researchers recently revisited that experiment and found that it contains a definitive quantitative indicator of quality that is consistent across all of the sites. The MIT researchers also found that the unpredictability of the results may have as much to do with the way the test sites were organized as with social influence.
The MIT researchers developed a mathematical model that predicts the experimental results with high accuracy. They found the percentage of customers who would download a given song after sampling it was consistent across all the sites, and the difference in download totals was due to the decision to sample the song in the first place. "The model that they propose does a good job of providing insight into what’s happening in the experiment," says Princeton University professor Matthew Sagalnik.
From MIT News
View Full Article
Abstracts Copyright © 2012 Information Inc. 
, Bethesda, Maryland, USA
No entries found