You are correct that only a little bit of A'A is used for any given user's
recommendation, but it is the columns that correspond to their history, not
the rows. This doesn't matter much to get this wrong if A'A is still
symmetrical, but getting it right can make your code less confusing for
others and if you limit the number of nonzero elements in each row of A'A,
then it becomes important to keep track of which is which.
On Fri, Dec 4, 2009 at 12:57 AM, Gökhan Çapan <gkhncpn@gmail.com> wrote:
> ith row of the matrix A'A contains all items and their similarity degrees
> to
> the item that is represented at ith column of the matrix A.
> I guess it is enough using only a subset of A'A at the final step, that is,
> the rows which represent the items that are in active user's history.
> btw, I also want to contribute to that implementation, if we can decide the
> algorithm.
>
> 
Ted Dunning, CTO
DeepDyve
