A graphical model formulation of collaborative filtering neighbourhood methods with fast maximum entropy training
dc.contributor.author | Defazio, Aaron | |
dc.contributor.author | Caetano, Tiberio | |
dc.coverage.spatial | Edinburgh UK | |
dc.date.accessioned | 2015-12-10T23:31:35Z | |
dc.date.created | June 26-July 1 2012 | |
dc.date.issued | 2012 | |
dc.date.updated | 2016-02-24T08:50:59Z | |
dc.description.abstract | 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. | |
dc.identifier.isbn | 9781450312851 | |
dc.identifier.uri | http://hdl.handle.net/1885/68699 | |
dc.publisher | Conference Organising Committee | |
dc.relation.ispartofseries | International Conference on Machine Learning (ICML 2012) | |
dc.source | Proceedings of the 29th International Conference on Machine Learning, ICML 2012 | |
dc.subject | Keywords: Bethe approximation; Classical approach; Collaborative filtering; Data sets; Gradient ascent; GraphicaL model; Maximum entropy; Maximum likelihood approaches; Neighbourhood; Nonlocal; Orders of magnitude; Prediction rules; Sufficient statistics; Training | |
dc.title | A graphical model formulation of collaborative filtering neighbourhood methods with fast maximum entropy training | |
dc.type | Conference paper | |
local.bibliographicCitation.lastpage | 272 | |
local.bibliographicCitation.startpage | 265 | |
local.contributor.affiliation | Defazio, Aaron, College of Engineering and Computer Science, ANU | |
local.contributor.affiliation | Caetano, Tiberio, College of Engineering and Computer Science, ANU | |
local.contributor.authoremail | u4406979@anu.edu.au | |
local.contributor.authoruid | Defazio, Aaron, u4406979 | |
local.contributor.authoruid | Caetano, Tiberio, u4590840 | |
local.description.embargo | 2037-12-31 | |
local.description.notes | Imported from ARIES | |
local.description.refereed | Yes | |
local.identifier.absfor | 080104 - Computer Vision | |
local.identifier.absfor | 100503 - Computer Communications Networks | |
local.identifier.ariespublication | f5625xPUB1802 | |
local.identifier.scopusID | 2-s2.0-84867114720 | |
local.identifier.uidSubmittedBy | f5625 | |
local.type.status | Published Version |
Downloads
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- 01_Defazio_A_graphical_model_formulation_2012.pdf
- Size:
- 396.76 KB
- Format:
- Adobe Portable Document Format