mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Owen <sro...@gmail.com>
Subject Re: Mahout "explain" process
Date Thu, 08 Sep 2011 20:29:01 GMT
I think he or she is just referring to the method
ItemBasedRecommender.recommendedBecause(). This is as close to an "explain"
operation as there is in the API.

In reality recommendations are a function of all data. In practice, what you
are asking for is the items most similar to well-liked items.
Recommendations are a function of more than this, but you could say these
are among the most influential reasons.

Really you want something like UserBasedRecommender.recommendedBecause()
since you're dealing in similar users, but that doesn't exist for no really
good reason. You could implement this and make a patch, just by imitating
the existing recommendedBecause() method.

Sean

On Thu, Sep 8, 2011 at 5:10 PM, Klokie Grossfeld <mahout@klokie.com> wrote:

> Hi, I've just started working with the Recommender API for Drupal,
> which integrates with Mahout. I'm reading up on Mahout, but I haven't
> figured out how to determine which content has been used to compute a
> given positive recommendation, i.e. how to obtain which nodes were
> used to compute an index of similarity. For example, I would like to
> display to the end user some text on a page they rated highly, like
> "You may also like these other nodes, since two other people [with
> similar affinities] also rated them highly".
>
> The developer of the Recommender API modules pointed me toward the
> Mahout "explain" process, but I can't seem to find any information on
> this. Could someone please point me in the right direction?
>
> thanks
> Klokie
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message