In this thesis, we give methods for time-aware recommendation, based on our paper [Pálovics, Benczúr, Kocsis, Kiss, Frigó, RecSys 2014] and additional experiments with more data sets and distributed implementations.
Our results have the potential of exploiting immediate temporal influences between users in a social media service.
We provide new models and evaluation
measures for real time recommendations with very strong temporal aspects by
introducing an online DCG evaluation metric.
Our experiments are carried over the two-year "scrobble" history of 70,000 Last.fm users and the public MovieLens data.
We consider the distributed implementation of our methods in large fully distributed P2P systems, in a
robust and scalable manner. We assume that the matrix to be
approximated is stored in a large network where each node
knows one row of the matrix (personal attributes, documents,
media ratings, etc.). In our P2P model, we do not allow this
personal information to leave the node, yet we want the nodes to
collaboratively train a recommender system. Methods applied in large scale
distributed systems such as synchronized parallel gradient search
or distributed iterative methods are not preferable in our system
model due to their requirements of synchronized rounds or their
inherent issues with load balancing. Our approach overcomes
these limitations with the help of a distributed stochastic gradient
search in which the personal part of the decomposition remains
local, and the global part (e.g., movie features) converges at all
nodes to the correct value.
Our main contribution is the online and distributed implementation of the factor models in the recommender frameworks. Using our implementation, we investigated the performance of online factorization algorithms and compared them with traditionally trained models.