How to predict who will win American Idol using complex software

January 13, 2009 by Lisa Damast
Filed under: Software, Uncategorized, Web2.0

Prof. Yuval Shavitt

Not happy with the prediction software of TiVo, Google Trends or DialIdol.com? According to the website of a Fox News affiliate, Israeli researchers from Tel Aviv University (TAU) have developed software that

“uses a mathematical formula to sort music requests logged by the Gnutella peer-to-peer file-sharing network to predict the next pop star.”

The researchers came up with the geographical formula – which has had a 30-50% rate of success so far – after realizing that

“those artists who eventually made it big on the national level first had a huge number of user queries in their local region, even when they had zero queries from elsewhere in the United States.”

Record companies can find this software useful as an added measure to determine which new signings, half of whom fail, will have the most potential. Companies in other fields can apply the formula to other entertainment areas including television programs and video and animation clips – including YouTube, Metacafe, AniBoom and other similar sites.

This software was developed by Professor Yuval Shavitt and a student of his as part of the DIMES project which aims “to study the structure and topology of the Internet.”

Share and Enjoy: