In contrast to
the previous works, this method is developed specially
for aspect model in recommender system. The proposed
method selects items for querying which are most effective
to improve the user latent parameters in aspect
model.
• The proposed approach results in drastically reduced user
waiting times (one order of magnitude). The reason is that
this method takes into account the difference between
classic active learning and active learning for recommender
system. In the recommender system context, each
item has already been rated by training users while in
classic active learning there is not training user. Considering
this difference, we can find new algorithms which
rely on this additional information instead of complicated
computations.
The rest of this paper is organized as follows: in section 2,
the related works are reviewed. Active learning for aspect is
explained in section 3. Then the proposed method will be
described in section 4, followed by experimental result in
section 5. Finally the conclusion is stated in section 6.