[A2k] The Netflix prize, about to be won
James Love
james.love@keionline.org
Sat Jun 27 12:35:02 2009
http://www.keionline.org/blogs/2009/06/27/the-netflix-prize/
The Netflix prize, about to be won
Knowledge Ecology Notes
By James Love, on June 27th, 2009
Apparently at least one entrant has qualified to win the $1 million
Netflix grand prize. One can follow the progress of the teams on the
Netflix Leaderboard: http://www.netflixprize.com/leaderboard.
Netflix is company that lends movie DVDs by postal mail or streams
directly to a computing device, for an monthly subscription fee.
According to the company, the =E2=80=9CNetflix Prize seeks to substantially
improve the accuracy of predictions about how much someone is going to
love a movie based on their movie preferences. Improve it enough and you
win one (or more) Prizes. Winning the Netflix Prize improves our ability
to connect people to the movies they love.=E2=80=9D The contest began in Oc=
tober
2006, and features annual =E2=80=9Cprogress=E2=80=9D prizes of $50,000 (for=
the best
result achieved), and a grand prize, now apparently about to be won, of
$1 million, for improving the Netfix predictions of movies by 10
percent.
An interesting account of the competition, from some members of the
first team to qualify for the grand prize, was published in the May 2009
IEEE Spectrum.*
The detailed rules that are available here, include this summary.
Terms and Conditions in a Nutshell
* Contest begins October 2, 2006 and continues through at least
October 2, 2011.
* Contest is open to anyone, anywhere (except certain countries
listed below).
* You have to register to enter.
* Once you register and agree to these Rules, you=E2=80=99ll have a=
ccess
to the Contest training data and qualifying test sets.
* To qualify for the $1,000,000 Grand Prize, the accuracy of
your submitted predictions on the qualifying set must be at least 10%
better than the accuracy Cinematch can achieve on the same training data
set at the start of the Contest.
* To qualify for a year=E2=80=99s $50,000 Progress Prize the accura=
cy of
any of your submitted predictions that year must be less than or equal
to the accuracy value established by the judges the preceding year.
* To win and take home either prize, your qualifying submissions
must have the largest accuracy improvement verified by the Contest
judges, you must share your method with (and non-exclusively license it
to) Netflix, and you must describe to the world how you did it and why
it works.
The intellectual property rights for the =E2=80=9CWinning Algorithm=E2=80=
=9D are handled
as follows:
----
After qualifying for either the Grand or Progress Prize and being
verified by the Contest judges, as a condition to receiving either
Prize, the winning individual and/or team must grant to Netflix
(including its affiliates and subsidiaries, employees, agents, and
contractors), an irrevocable, royalty free, fully paid up, worldwide
non-exclusive license under the Participants=E2=80=99 copyrights, patents o=
r
other intellectual property rights in the winning algorithm (=E2=80=9DWinni=
ng
Algorithm=E2=80=9D) to reproduce, distribute, display, and create derivativ=
e
works from the Winning Algorithm and also to make, have made, use, sell,
offer for sale, and import products that would otherwise infringe the
Winning Algorithm. Except as encompassed in the concept of =E2=80=9Chave ma=
de=E2=80=9D,
this license will not include the right to grant further licenses or
sublicenses.
-----
The Netflix prize stimulated a large academic literature of data mining
techniques [such as these articles in Google Scholar], and reportedly
has attracted so far 49,251 contestants on 40,467 teams from 184
different countries.
In recent blog, Melody Hildebrandt offered this comment regarding the
incentives in the prize contest to collaborate:
-----
This is one of the best examples of crowdsourced innovation and
problem-solving out there. The winning team was initially 4 disparate
pairs or individuals who they realized that they had complementary
skills =E2=80=94 machine learning, computer science, engineering =E2=80=94 =
and decided
to collaborate. Throughout the competition, as the market leaderboard
tracked the top performers, teams would routinely share lessons learned.
Even with $1M at stake, the market can indeed be collaborative and come
to a better solution than a single company alone.
-----
*Bell, Robert M., Jim Bennett, Yehuda Koren, and Chris Volinsky. =E2=80=9CT=
he
Million Dollar Programming Prize: Netflix=E2=80=99s bounty for improving it=
s
movie-recommendation software is almost in the bag. Here is one team=E2=80=
=99s
account.=E2=80=9D IEEE Spectrum, May 2009.
http://www.spectrum.ieee.org/computing/software/the-million-dollar-programm=
ing-prize/0.
See also:
Buskirk, Eliot Van. =E2=80=9C$1 Million Netflix Prize So Close, They Can Ta=
ste
It.=E2=80=9D Wired.Com, June 17, 2009.
http://www.wired.com/epicenter/2009/06/1-million-netflix-prize-so-close-the=
y-can-taste-it/.
Hildebrandt, Melody. =E2=80=9CNetflix prize is (nearly) awarded! A model in
crowdsourcing.=E2=80=9D TransCapitalist, June 26, 2009.
http://www.transcapitalist.com/transcapitalist/2009/6/26/netflix-prize-is-n=
early-awarded-a-model-in-crowdsourcing.html.
Lohr, Steve. =E2=80=9CAnd the Winner of the $1 Million Netflix Prize (Proba=
bly)
Is =E2=80=A6.=E2=80=9D New York Times, June 26, 2009.
http://bits.blogs.nytimes.com/2009/06/26/and-the-winner-of-the-1-million-ne=
tflix-prize-probably-is/.
Thompson, Clive. =E2=80=9CIf You Liked This, You=E2=80=99re Sure to Love Th=
at.=E2=80=9D The New
York Times, November 23, 2008, sec. Magazine.
http://www.nytimes.com/2008/11/23/magazine/23Netflix-t.html.
For more background on innovation inducement prizes, see:
March 7, 2008. Selected Innovation Prizes and Reward Programs, KEI
Research Note 2008:1
(http://www.keionline.org/misc-docs/research_notes/kei_rn_2008_1.pdf)
An Annotated Bibliography of Scholarly and Technical Articles and Books
on Innovation Prizes, KEI Research Note 2008:2
http://www.zotero.org/groups/innovation_inducement_prizes
Prizes to Stimulate Innovation:
http://www.keionline.org/content/view/4/1/
--
James Love, Director, Knowledge Ecology International
http://www.keionline.org | mailto:james.love at keionline.org
Wk: +1.202.332.2671 | US Mobile +1.202.361.3040 | Geneva Mobile +41.76.413.=
6584