[Air-L] Research on collaborative filtering?

[Bernhard Rieder]

Hi Jason,

Duncan Watts (a physicist gone sociologist) has done a large scale 
experiment on the collective emergence of taste, not really via direct 
GroupLens-like technologies but a system based on voting (not that 
dissimilar, really): 
http://www.nytimes.com/2007/04/15/magazine/15wwlnidealab.t.html?_r=1&ref=magazine 


You will normally find quite a lot on usage experiments in certain CS 
papers (normally follow up surveys to show that one algorithm is more 
useful than another), just skip the formal discussion...

best,
Bernhard

-- 
Bernhard Rieder
Laboratoire Paragraphe
Université de Paris 8

++33 6 60 87 80 54
bernhard.rieder at univ-paris8.fr
http://bernhard.rieder.fr
http://thepoliticsofsystems.net

Jason Mittell wrote:
> Hi all - inspired by the excellent Clive Thompson article in today's NYT 
> on Netflix & collaborative filtering 
> (http://www.nytimes.com/2008/11/23/magazine/23Netflix-t.html ), I want 
> to bring a research project off the backburner about collaborative 
> filtering and taste cultures. Does anyone know of any research from a 
> more cultural/social perspective on CF? The research that's I've found 
> via searches is all about algorithms - since I don't want to build a 
> system, but rather study their use & significance, those articles are 
> less than useful (and utterly incomprehensible to my humanistic mind). 
> I'd appreciate any references about recommendation systems and 
> collectivized taste that you know of.
> 
> And on a related note, if other people are interested in the topic, 
> might we want to put together a panel for IR 10.0? If so, contact me 
> off-list.
> 
> Thanks,
> -Jason
>