On 2013-08-07 03:24, Ted Dunning wrote:
Yes. There are several approaches.
One of the most effective is rescoring. You use a performant
recommender
such as a search engine based recommender and then rescore the top few
hundred items using a more detailed model.
That's what I was thinking about. Currently I recommend a fixed number
of products. Those item should be used to take product details into
account.
This typically won't be fast enough if you have something like a
random
forest, but if your final targeting model is logistic regression, it
probably will be fast enough.
So usually I do need to train a custom model for each user
independently?
In any case, there are also tricks you can pull in the evaluation of
certain classes of models. For instance, with logistic regression,
you can
remove the link function (doesn't change ordering) and you can ignore
all
user specific features and weights (this doesn't change ordering
either).
This leaves you with a relatively small number of computations in the
form
of a sparse by dense dot product.
