schliessen

Filtern

 

Bibliotheken

Field-Aware Matrix Factorization for Recommender Systems

Predicting user response is one of the core machine learning tasks in recommender systems (RS). The matrix factorization (MF)-based model has been proved to be a useful tool to improve the performance of recommendation. Many existing matrix factorization-based models mainly rely on adding some side... Full description

Journal Title: IEEE Access 2018, Vol.6, pp.45690-45698
Main Author: Zhang, Zhiyuan
Other Authors: Liu, Yun , Zhang, Zhenjiang
Format: Electronic Article Electronic Article
Language: English
Subjects:
ID: E-ISSN: 2169-3536 ; DOI: 10.1109/ACCESS.2017.2787741
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: ieee_s8241366
title: Field-Aware Matrix Factorization for Recommender Systems
format: Article
creator:
  • Zhang, Zhiyuan
  • Liu, Yun
  • Zhang, Zhenjiang
subjects:
  • Sparse Matrices
  • Matrix Decomposition
  • Motion Pictures
  • Data Models
  • Probabilistic Logic
  • Frequency Modulation
  • Recommender Systems
  • Recommender Systems
  • Matrix Factorization
  • Machine Learning
  • Field
  • Engineering
ispartof: IEEE Access, 2018, Vol.6, pp.45690-45698
description: Predicting user response is one of the core machine learning tasks in recommender systems (RS). The matrix factorization (MF)-based model has been proved to be a useful tool to improve the performance of recommendation. Many existing matrix factorization-based models mainly rely on adding some side information into basic MF to enable the model to fully express the data. However, most of the side information is measured based on the statistics or empirical formula. Also, the latent features of side information cannot be deeply mined. In this paper, we focus on mining the influence of field information (useful side information) to improve the performance of prediction. Based on the MF framework, we propose a field-aware matrix factorization (FMF) model. In FMF, the interactions between user/item and field can be captured and learned in the latent vector spaces. We propose efficient implementations to train FMF. Then, we comprehensively analyze FMF and compare this model with the state-of-the-art models. The analysis of experiments on two large data sets demonstrates that our method is very useful in RS.
language: eng
source:
identifier: E-ISSN: 2169-3536 ; DOI: 10.1109/ACCESS.2017.2787741
fulltext: fulltext_linktorsrc
issn:
  • 2169-3536
  • 21693536
url: Link


@attributes
ID1459036238
RANK0.07
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
LOCALfalse
PrimoNMBib
record
control
sourcerecordid8241366
sourceidieee_s
recordidTN_ieee_s8241366
sourcesystemOther
dbid
097E
1AAJGR
2AASAJ
3ATWAV
4BEFXN
5BFFAM
6BGNUA
7BKEBE
8BPEOZ
9ESBDL
10IPLJI
11OCL
12RIA
13RIE
display
typearticle
titleField-Aware Matrix Factorization for Recommender Systems
creatorZhang, Zhiyuan ; Liu, Yun ; Zhang, Zhenjiang
ispartofIEEE Access, 2018, Vol.6, pp.45690-45698
identifierE-ISSN: 2169-3536 ; DOI: 10.1109/ACCESS.2017.2787741
subjectSparse Matrices ; Matrix Decomposition ; Motion Pictures ; Data Models ; Probabilistic Logic ; Frequency Modulation ; Recommender Systems ; Recommender Systems ; Matrix Factorization ; Machine Learning ; Field ; Engineering
descriptionPredicting user response is one of the core machine learning tasks in recommender systems (RS). The matrix factorization (MF)-based model has been proved to be a useful tool to improve the performance of recommendation. Many existing matrix factorization-based models mainly rely on adding some side information into basic MF to enable the model to fully express the data. However, most of the side information is measured based on the statistics or empirical formula. Also, the latent features of side information cannot be deeply mined. In this paper, we focus on mining the influence of field information (useful side information) to improve the performance of prediction. Based on the MF framework, we propose a field-aware matrix factorization (FMF) model. In FMF, the interactions between user/item and field can be captured and learned in the latent vector spaces. We propose efficient implementations to train FMF. Then, we comprehensively analyze FMF and compare this model with the state-of-the-art models. The analysis of experiments on two large data sets demonstrates that our method is very useful in RS.
languageeng
oafree_for_read
source
version2
lds50peer_reviewed
links
openurl$$Topenurl_article
openurlfulltext$$Topenurlfull_article
linktorsrc$$Uhttps://ieeexplore.ieee.org/document/8241366$$EView_full_text_in_IEEE_Xplore
search
creatorcontrib
0Zhang, Zhiyuan
1Liu, Yun
2Zhang, Zhenjiang
titleField-Aware Matrix Factorization for Recommender Systems
description

Predicting user response is one of the core machine learning tasks in recommender systems (RS). The matrix factorization (MF)-based model has been proved to be a useful tool to improve the performance of recommendation. Many existing matrix factorization-based models mainly rely on adding some side information into basic MF to enable the model to fully express the data. However, most of the side information is measured based on the statistics or empirical formula. Also, the latent features of side information cannot be deeply mined. In this paper, we focus on mining the influence of field information (useful side information) to improve the performance of prediction. Based on the MF framework, we propose a field-aware matrix factorization (FMF) model. In FMF, the interactions between user/item and field can be captured and learned in the latent vector spaces. We propose efficient implementations to train FMF. Then, we comprehensively analyze FMF and compare this model with the state-of-the-art models. The analysis of experiments on two large data sets demonstrates that our method is very useful in RS.

subject
0Sparse Matrices
1Matrix Decomposition
2Motion Pictures
3Data Models
4Probabilistic Logic
5Frequency Modulation
6Recommender Systems
7Matrix Factorization
8Machine Learning
9Field
10Engineering
general
0English
1IEEE
210.1109/ACCESS.2017.2787741
3IEEE Xplore
4IEEE Journals & Magazines 
5IEEE Conference Publications
6IEEE Journals & Magazines
sourceidieee_s
recordidieee_s8241366
issn
02169-3536
121693536
rsrctypearticle
creationdate2018
addtitleIEEE Access
searchscope
0ieee_full
1ieee13
2ieee32
3ieee14
4ieee44
5ieee45
6ieee46
7ieee47
8ieee48
9ieee2
10ieee42
11ieee40
12ieee12
13ieee8
scope
0ieee_full
1ieee13
2ieee32
3ieee14
4ieee44
5ieee45
6ieee46
7ieee47
8ieee48
9ieee2
10ieee42
11ieee40
12ieee12
13ieee8
lsr45$$EView_full_text_in_IEEE_Xplore
tmp01
0IEEE Journals & Magazines 
1IEEE Xplore
2IEEE Conference Publications
3IEEE Journals & Magazines
tmp02
097E
1AAJGR
2AASAJ
3ATWAV
4BEFXN
5BFFAM
6BGNUA
7BKEBE
8BPEOZ
9ESBDL
10IPLJI
11OCL
12RIA
13RIE
orcidid0000-0002-2861-0316
startdate20180101
enddate20181231
lsr40IEEE Access, 2018, Vol.6, pp.45690-45698
doi10.1109/ACCESS.2017.2787741
citationpf 45690 pt 45698 vol 6
lsr30VSR-Enriched:[issue]
sort
titleField-Aware Matrix Factorization for Recommender Systems
authorZhang, Zhiyuan ; Liu, Yun ; Zhang, Zhenjiang
creationdate20180000
lso0120180000
facets
frbrgroupid4240926650927272308
frbrtype5
newrecords20190131
languageeng
topic
0Sparse Matrices
1Matrix Decomposition
2Motion Pictures
3Data Models
4Probabilistic Logic
5Frequency Modulation
6Recommender Systems
7Matrix Factorization
8Machine Learning
9Field
10Engineering
collectionIEEE Xplore
prefilterarticles
rsrctypearticles
creatorcontrib
0Zhang, Zhiyuan
1Liu, Yun
2Zhang, Zhenjiang
jtitleIEEE Access
creationdate2018
toplevelpeer_reviewed
delivery
delcategoryRemote Search Resource
fulltextfulltext_linktorsrc
addata
orcidid0000-0002-2861-0316
aulast
0Zhang
1Liu
aufirst
0Zhiyuan
1Yun
2Zhenjiang
auinitZ
auinit1Z
au
0Zhang, Zhiyuan
1Liu, Yun
2Zhang, Zhenjiang
atitleField-Aware Matrix Factorization for Recommender Systems
jtitleIEEE Access
date2018
risdate2018
volume6
spage45690
epage45698
pages45690-45698
eissn2169-3536
formatjournal
genrearticle
ristypeJOUR
abstract

Predicting user response is one of the core machine learning tasks in recommender systems (RS). The matrix factorization (MF)-based model has been proved to be a useful tool to improve the performance of recommendation. Many existing matrix factorization-based models mainly rely on adding some side information into basic MF to enable the model to fully express the data. However, most of the side information is measured based on the statistics or empirical formula. Also, the latent features of side information cannot be deeply mined. In this paper, we focus on mining the influence of field information (useful side information) to improve the performance of prediction. Based on the MF framework, we propose a field-aware matrix factorization (FMF) model. In FMF, the interactions between user/item and field can be captured and learned in the latent vector spaces. We propose efficient implementations to train FMF. Then, we comprehensively analyze FMF and compare this model with the state-of-the-art models. The analysis of experiments on two large data sets demonstrates that our method is very useful in RS.

pubIEEE
doi10.1109/ACCESS.2017.2787741
oafree_for_read
issue99