Learning Centre

On two-way grouping by one-way topic models

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.author Savia, Eerika
dc.contributor.author Puolamäki, Kai
dc.contributor.author Kaski, Samuel
dc.date.accessioned 2011-11-28T13:22:52Z
dc.date.available 2011-11-28T13:22:52Z
dc.date.issued 2009
dc.identifier.isbn 978-951-22-9947-8
dc.identifier.issn 1797-5042
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/895
dc.description.abstract We tackle the problem of new users or documents in collaborative filtering. Generalization over users by grouping them into user groups is beneficial when a rating is to be predicted for a relatively new document having only few observed ratings. The same applies for documents in the case of new users. We have shown earlier that if there are both new users and new documents, two-way generalization becomes necessary, and introduced a probabilistic Two-Way Model for the task. The task of finding a two-way grouping is a non-trivial combinatorial problem, which makes it computationally difficult. We suggest approximating the Two-Way Model with two URP models; one that groups users and one that groups documents. Their two predictions are combined using a product of experts model. This combination of two one-way models achieves even better prediction performance than the original Two-Way Model. This article contains the full technical details of the conference article [22]. en
dc.format.extent 29
dc.format.mimetype application/pdf
dc.language.iso en en
dc.publisher Helsinki University of Technology en
dc.publisher Teknillinen korkeakoulu fi
dc.relation.ispartofseries TKK reports in information and computer science en
dc.relation.ispartofseries 15 en
dc.subject.other Computer science en
dc.title On two-way grouping by one-way topic models en
dc.type D4 Julkaistu kehittämis- tai tutkimusraportti taikka -selvitys fi
dc.contributor.school Faculty of Information and Natural Sciences en
dc.contributor.school Informaatio- ja luonnontieteiden tiedekunta fi
dc.contributor.department Department of Information and Computer Science en
dc.contributor.department Tietojenkäsittelytieteen laitos fi
dc.subject.keyword latent topic model en
dc.subject.keyword collaborative filtering en
dc.subject.keyword cold-start problem en
dc.subject.keyword product of experts en
dc.identifier.urn urn:nbn:fi:tkk-013005
dc.type.dcmitype text en

Files in this item

This item appears in the following Collection(s)

Show simple item record

Search archive

Advanced Search

article-iconSubmit a publication