Thank you Ted, obviously you guys had been thinking about these for alot longer time than I have. I will try these different approaches to see what comes out. Its definitely not intuitive to me :-).
Also I miss-spoke about the dataset, I meant 120K unique users for 1200+ items, number of ratings are in the millions. However this is one of the smaller datasets (for number of items) we have, although it will be the first one to get feedback data on. On Wed, Apr 28, 2010 at 3:19 PM, Ted Dunning <ted.dunn...@gmail.com> wrote: > This is a small data set, but it would be of real value for testing of > these > kinds of problems. > > 2010/4/28 Tolga Oral <tolga.o...@gmail.com> > > > (Chen) If interested, I could probably provide our dataset (anonymized > for > > users and items). Our system is providing recommendations for multiple > > silos > > (different set of unrelated items), specific silo we will be getting > > negative feedback have 120K+ user ratings for 1200+ items. Different > silos > > have different characteristics for example recommendation for urls has > alot > > more items then this specific silo. > > >
