You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by Chris Schilling <ch...@cellixis.com> on 2011/02/14 19:38:47 UTC
item similarities
Hi,
I am sure this is very straight-foward, but I could not find what I am looking for in MIA or in the docs. I am looking for an API to return the top most similar items in item based CF. I see that I could use the itemsimilarity command line option to calculate the similarities. The thing is, I do not really need to map-reduce the calculation. At the stage I am at, I could do it in memory sequentially. So, is there an API for this and then how do I get the list of the top n similar items.
Thanks,
Chris S.
Re: item similarities
Posted by Chris Schilling <ch...@cellixis.com>.
Thanks Sean!
I was ignorantly looking in the interface ItemSimilarity :) I see it now...
On Feb 14, 2011, at 11:15 AM, Sean Owen wrote:
> Yes just call ItemBasedRecommender.mostSimilarItems()
>
> On Mon, Feb 14, 2011 at 6:38 PM, Chris Schilling <ch...@cellixis.com> wrote:
>> Hi,
>>
>> I am sure this is very straight-foward, but I could not find what I am looking for in MIA or in the docs. I am looking for an API to return the top most similar items in item based CF. I see that I could use the itemsimilarity command line option to calculate the similarities. The thing is, I do not really need to map-reduce the calculation. At the stage I am at, I could do it in memory sequentially. So, is there an API for this and then how do I get the list of the top n similar items.
>>
>> Thanks,
>> Chris S.
>>
>>
Re: item similarities
Posted by Sean Owen <sr...@gmail.com>.
Yes just call ItemBasedRecommender.mostSimilarItems()
On Mon, Feb 14, 2011 at 6:38 PM, Chris Schilling <ch...@cellixis.com> wrote:
> Hi,
>
> I am sure this is very straight-foward, but I could not find what I am looking for in MIA or in the docs. I am looking for an API to return the top most similar items in item based CF. I see that I could use the itemsimilarity command line option to calculate the similarities. The thing is, I do not really need to map-reduce the calculation. At the stage I am at, I could do it in memory sequentially. So, is there an API for this and then how do I get the list of the top n similar items.
>
> Thanks,
> Chris S.
>
>