Information filtering based on corrected redundancy-eliminating mass diffusion

Methods used in information filtering and recommendation often rely on quantifying the similarity between objects or users. The used similarity metrics often suffer from similarity redundancies arising from correlations between objects' attributes. Based on an unweighted undirected object-user...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:PloS one Ročník 12; číslo 7; s. e0181402
Hlavní autoři: Zhu, Xuzhen, Yang, Yujie, Chen, Guilin, Medo, Matus, Tian, Hui, Cai, Shi-Min
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States Public Library of Science 27.07.2017
Public Library of Science (PLoS)
Témata:
Accuracy
Algorithms
Area Under Curve
Automatic classification
Automatic indexing
Biology and Life Sciences
Collaboration
Communications networks
Computer and Information Sciences
Computer Communication Networks
Customers
Databases as Topic
Diffusion
Electronic commerce
Filtration
Methods
Physical Sciences
Recommender systems
Redundancy
Research and Analysis Methods
Similarity
Social Sciences
Sparsity
Beijing China
China
ISSN:1932-6203, 1932-6203
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Methods used in information filtering and recommendation often rely on quantifying the similarity between objects or users. The used similarity metrics often suffer from similarity redundancies arising from correlations between objects' attributes. Based on an unweighted undirected object-user bipartite network, we propose a Corrected Redundancy-Eliminating similarity index (CRE) which is based on a spreading process on the network. Extensive experiments on three benchmark data sets-Movilens, Netflix and Amazon-show that when used in recommendation, the CRE yields significant improvements in terms of recommendation accuracy and diversity. A detailed analysis is presented to unveil the origins of the observed differences between the CRE and mainstream similarity indices.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Competing Interests: The authors have declared that no competing interests exist.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0181402