Talk:Netflix Prize/Archives/2012

Latest comment: 14 years ago by Araes in topic Dataset Size


No need for the play-by-play

I think the Progress section should just mention the perennial leaders, with a final "as of" paragraph. Ideally, we'd have a graph of the best scores ... —johndburger 01:37, 22 August 2007 (UTC)

Like this? Too bad it's from BellKor, so we can't use it. -kslays (talk) 02:43, 29 January 2008 (UTC)

Privacy

There are some privacy concerns over the dataset that Netflix released to contestants. Apparently, the "data perturbation" doesn't quite do the trick. See this source. Copana2002 (talk) 16:07, 27 March 2009 (UTC)

Over?

"The Netflix prize has been won", Marginal Revolution, June 26, 2009.  Skomorokh  21:03, 26 June 2009 (UTC)

Dataset Size

The initial description of the dataset size is misleading the way it currently is stated at "100,480,507 million ratings" This implies that there are in fact 100 trillion data points, whereas there are only 100 million. Changing to 100,480,507 ratings, and source [2] would still be valid for this. Araes (talk) 22:17, 29 January 2010 (UTC)