TY - GEN

T1 - A graphical model formulation of collaborative filtering neighbourhood methods with fast maximum entropy training

AU - Defazio, Aaron J.

AU - Caetano, Tibério S.

PY - 2012

Y1 - 2012

N2 - Item neighbourhood methods for collaborative filtering learn a weighted graph over the set of items, where each item is connected to those it is most similar to. The prediction of a user's rating on an item is then given by that rating of neighbouring items, weighted by their similarity. This paper presents a new neighbourhood approach which we call item fields, whereby an undirected graphical model is formed over the item graph. The resulting prediction rule is a simple generalization of the classical approaches, which takes into account non-local information in the graph, allowing its best results to be obtained when using drastically fewer edges than other neighbourhood approaches. A fast approximate maximum entropy training method based on the Bethe approximation is presented, which uses a simple gradient ascent procedure. When using precomputed sufficient statistics on the Movielens datasets, our method is faster than maximum likelihood approaches by two orders of magnitude.

AB - Item neighbourhood methods for collaborative filtering learn a weighted graph over the set of items, where each item is connected to those it is most similar to. The prediction of a user's rating on an item is then given by that rating of neighbouring items, weighted by their similarity. This paper presents a new neighbourhood approach which we call item fields, whereby an undirected graphical model is formed over the item graph. The resulting prediction rule is a simple generalization of the classical approaches, which takes into account non-local information in the graph, allowing its best results to be obtained when using drastically fewer edges than other neighbourhood approaches. A fast approximate maximum entropy training method based on the Bethe approximation is presented, which uses a simple gradient ascent procedure. When using precomputed sufficient statistics on the Movielens datasets, our method is faster than maximum likelihood approaches by two orders of magnitude.

UR - http://www.scopus.com/inward/record.url?scp=84867114720&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781450312851

T3 - Proceedings of the 29th International Conference on Machine Learning, ICML 2012

SP - 265

EP - 272

BT - Proceedings of the 29th International Conference on Machine Learning, ICML 2012

T2 - 29th International Conference on Machine Learning, ICML 2012

Y2 - 26 June 2012 through 1 July 2012

ER -