I want to get the predicted rating predicted rating for users that I
specify for specific items (which will vary from user to user)

E.g.

User1 : I want to get predicted rating for Item_67
user1: i Want to get predicted rating for Item_23
user1: i Want to get predicted rating for Item_456

user2: i Want to get predicted rating for Item_12

user97: i Want to get predicted rating for Item_34
user97: i Want to get predicted rating for Item_11
user97: i Want to get predicted rating for Item_ 45


I want to specify user and some items corresponding to that user. Then I
want to get a predicted rating for those items for that user.

This is not possible using the fully distributed algorithm in mahout.My
question was: can i use the pseudo distributed algo to do this? is there
some built in function like Recommender.estimatePreference?



On Tue, Nov 15, 2011 at 4:42 PM, Sean Owen <[email protected]> wrote:

> Use what, your data? surely. You can run many instances of a
> non-disributed algorithms on Hadoop this way. It does not really lose
> performance. But you don't have to code anything either way. And
> unless you need to move to Hadoop at all, don't. I'm not sure what
> you're asking.
>
> On Tue, Nov 15, 2011 at 3:53 AM, Akshay Jain <[email protected]> wrote:
> > Is there any way I can use this through the Pseudo-distributed algo? I
> dont
> > mind loosing some performance if I dont have to code the entire thing
> > (which I do not know how to)
> >
> > On Wed, Nov 9, 2011 at 4:36 PM, Akshay Jain <[email protected]> wrote:
> >
> >> @Sean, I am just testing with a small dataset. I have some large
> datasets
> >> which I am planning to use on Hadoop.
> >>
> >> Thanks.
> >>
> >> Akshay
>

Reply via email to