Distributed Bayesian Probabilistic Matrix Factorization

Using the matrix factorization technique in machine learning is very common mainly in areas like rec-ommender systems. Despite its high prediction accuracy and its ability to avoid over-fitting of the data, the Bayesian Probabilistic Matrix Factorization algorithm (BPMF) has not been widely used on...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Procedia computer science Jg. 108; S. 1030 - 1039
Hauptverfasser: Vander Aa, Tom, Chakroun, Imen, Haber, Tom
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 2017
Schlagworte:
ISSN:1877-0509, 1877-0509
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Using the matrix factorization technique in machine learning is very common mainly in areas like rec-ommender systems. Despite its high prediction accuracy and its ability to avoid over-fitting of the data, the Bayesian Probabilistic Matrix Factorization algorithm (BPMF) has not been widely used on large scale data because of the prohibitive cost. In this paper, we propose a distributed high-performance parallel implementation of the BPMF using Gibbs sampling on shared and distributed architectures. We show by using efficient load balancing using work stealing on a single node, and by using asynchronous communication in the distributed version we beat state of the art implementations.
AbstractList Using the matrix factorization technique in machine learning is very common mainly in areas like rec-ommender systems. Despite its high prediction accuracy and its ability to avoid over-fitting of the data, the Bayesian Probabilistic Matrix Factorization algorithm (BPMF) has not been widely used on large scale data because of the prohibitive cost. In this paper, we propose a distributed high-performance parallel implementation of the BPMF using Gibbs sampling on shared and distributed architectures. We show by using efficient load balancing using work stealing on a single node, and by using asynchronous communication in the distributed version we beat state of the art implementations.
Author Chakroun, Imen
Vander Aa, Tom
Haber, Tom
Author_xml – sequence: 1
  givenname: Tom
  surname: Vander Aa
  fullname: Vander Aa, Tom
  email: tom.vanderaa@imec.be
  organization: Exascience Lab, Imec, Kapeldreef 75, B-3001 Leuven, Belgium
– sequence: 2
  givenname: Imen
  surname: Chakroun
  fullname: Chakroun, Imen
  email: imen.chakroun@imec.be
  organization: Exascience Lab, Imec, Kapeldreef 75, B-3001 Leuven, Belgium
– sequence: 3
  givenname: Tom
  surname: Haber
  fullname: Haber, Tom
  email: tom.haber@uhasselt.be
  organization: Expertise Centre for Digital Media, Wetenschapspark 2, 3590 Diepenbeek, Belgium
BookMark eNqFj8FOwzAMhiM0JMbYE3DpC7Q4Tds0Bw4wGCANwQHOUeImUqbRTklAjKdftnFAHMCX35L9Wf5OyagfekPIOYWCAm0ulsXaDxiKEigvoC4AxBEZ05bzHGoQox_9CZmGsIRUrG0F5WPCb1yI3un3aLrsWm1McKrPnv2glXarNHOYPaq08ZnNFcbBuy8V3dCfkWOrVsFMv3NCXue3L7P7fPF09zC7WuTIqjbmWDZam051wvCKC0BgrAENtmyQMltZq21tS9FiVQuDjU5R15YyWjbcNMgmhB3uoh9C8MbKtXdvym8kBbnTl0u515c7fQm1TPqJEr8odHH_d_TKrf5hLw-sSVofzngZ0JkeTee8wSi7wf3JbwFMcHo7
CitedBy_id crossref_primary_10_1007_s10115_024_02315_z
crossref_primary_10_1007_s10994_019_05778_2
crossref_primary_10_3390_electronics11060878
crossref_primary_10_1007_s10994_019_05867_2
crossref_primary_10_1002_cpe_7879
crossref_primary_10_1186_s40537_021_00425_x
crossref_primary_10_1016_j_procs_2020_03_341
Cites_doi 10.1016/j.dss.2013.04.002
10.1137/1.9781611971408
10.1137/S0036144502409019
10.1145/2020408.2020426
10.1016/j.ijforecast.2006.03.001
10.1093/nar/gkt1031
10.1109/MC.2009.263
10.1145/2827872
10.1145/2568088.2576761
10.1145/2783258.2783373
10.1007/978-3-540-68880-8_32
10.1109/CLUSTER.2016.13
10.1145/1390156.1390267
ContentType Journal Article
Copyright 2017
Copyright_xml – notice: 2017
DBID 6I.
AAFTH
AAYXX
CITATION
DOI 10.1016/j.procs.2017.05.009
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 1877-0509
EndPage 1039
ExternalDocumentID 10_1016_j_procs_2017_05_009
S187705091730501X
GroupedDBID --K
0R~
0SF
1B1
457
5VS
6I.
71M
AACTN
AAEDT
AAEDW
AAFTH
AAIKJ
AALRI
AAQFI
AAXUO
ABMAC
ACGFS
ADBBV
ADEZE
AEXQZ
AFTJW
AGHFR
AITUG
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
E3Z
EBS
EJD
EP3
FDB
FNPLU
HZ~
IXB
KQ8
M41
M~E
NCXOZ
O-L
O9-
OK1
P2P
RIG
ROL
SES
SSZ
9DU
AAYWO
AAYXX
ABWVN
ACRPL
ACVFH
ADCNI
ADNMO
ADVLN
AEUPX
AFPUW
AIGII
AKBMS
AKRWK
AKYEP
CITATION
~HD
ID FETCH-LOGICAL-c348t-c26bbedad9e74790c03360b0f26c13f4ffbf5f298c459ec6b45955f131267e6c3
ISICitedReferencesCount 11
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000404959000104&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1877-0509
IngestDate Tue Nov 18 22:35:07 EST 2025
Sat Nov 29 07:51:49 EST 2025
Wed May 17 00:58:04 EDT 2023
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords multi-core
Collaborative filtering
PGAS
Probabilistic matrix factorization algorithm
Machine learning
Language English
License This is an open access article under the CC BY-NC-ND license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c348t-c26bbedad9e74790c03360b0f26c13f4ffbf5f298c459ec6b45955f131267e6c3
OpenAccessLink https://dx.doi.org/10.1016/j.procs.2017.05.009
PageCount 10
ParticipantIDs crossref_primary_10_1016_j_procs_2017_05_009
crossref_citationtrail_10_1016_j_procs_2017_05_009
elsevier_sciencedirect_doi_10_1016_j_procs_2017_05_009
PublicationCentury 2000
PublicationDate 2017
2017-00-00
PublicationDateYYYYMMDD 2017-01-01
PublicationDate_xml – year: 2017
  text: 2017
PublicationDecade 2010
PublicationTitle Procedia computer science
PublicationYear 2017
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Yong Guo, Ana Lucia Varbanescu, Alexandru Iosup, Claudio Martella, and Theodore L. Willke. Benchmarking graph-processing platforms: A vision. In Proceedings of the 5th ACM/SPEC International Conference on Performance Engineering, ICPE ‘14, pages 289–292, New York, NY, USA, 2014. ACM.
Ruslan Salakhutdinov and Andriy Mnih. Bayesian probabilistic matrix factorization using Markov chain Monte Carlo. In Proceedings of the International Conference on Machine Learning, volume 25, 2008.
Gilbert W. Stewart. Matrix algorithms. Volume 1. , Basic decompositions. SIAM (Society for industrial and applied mathematics), Philadelphia (Pa.), 1998.
Chakroun, Aa, De Fraine, Haber, Wuyts, De Meuter (bib0003) 2015
2010.
Sungjin Ahn, Anoop Korattikara Balan, Nathan Liu, Suju Rajan, and Max Welling. Large-scale distributed bayesian matrix factorization using stochastic gradient MCMC. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2015.
Daniel Grünewald and Christian Simmendinger. The GASPI API specification and its implementation GPI 2.0. In 7th International Conference on PGAS Programming Models, volume 243, 2013.
Virtanen, Klami, Khan, Kaski (bib0021) 2012
Koren, Bell, Volinsky (bib0011) Aug 2009; 42
Rainer Gemulla, Erik Nijkamp, Peter J. Haas, and Yannis Sismanis. Large-scale matrix factorization with distributed stochastic gradient descent. In Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ‘11, pages 69–77, New York, NY, USA, 2011. ACM.
Harper, Konstan (bib0008) December 2015; 5
Simm, Arany, Zakeri, Haber, Wegner, Chupakhin, Hugo, Yves (bib0017) 2015
Bento, Gaulton, Hersey, Bellis, Chambers, Davies, Kruger, Light, Mak, McGlinchey, Nowotka, Papadatos, Santos, Overington (bib0002) 2014; 42
MPI Forum. MPI: a message-passing interface standard. Technical report, Knoxville, TN, USA, 1994.
OpenMP Architecture Review Board. OpenMP application program interface version 3.0, May 2008.
Tom Vander Aa, Imen Chakroun, and Tom Haber. Distributed bayesian probabilistic matrix factorization. In 2016 IEEE International Conference on Cluster Computing (CLUSTER), pages 346–349, Sept 2016.
Gaël Guennebaud, Benoît Jacob, et al. Eigen v3.
Liua, Caihua (bib0010) June 2013; 55
Vastenhouw, Bisseling (bib0020) 2005; 47
Liu, Wu, Liu (bib0012) 2013; 55
Hyndman, Koehler (bib0009) 2006; 22
James Reinders. Intel Threading Building Blocks. O’Reilly & Associates, Inc., Sebastopol, CA, USA, first edition, 2007.
Yunhong Zhou, Dennis Wilkinson, Robert Schreiber, and Rong Pan. Large-scale parallel collaborative filtering for the Netflix prize. In Proceedings of the 4th International Conference on Algorithmic Aspects in Information and Management, AAIM ‘08, pages 337–348, Berlin, Heidelberg. Springer-Verlag.
Virtanen (10.1016/j.procs.2017.05.009_bib0021) 2012
Liu (10.1016/j.procs.2017.05.009_bib0012) 2013; 55
Hyndman (10.1016/j.procs.2017.05.009_bib0009) 2006; 22
Chakroun (10.1016/j.procs.2017.05.009_bib0003) 2015
Koren (10.1016/j.procs.2017.05.009_bib0011) 2009; 42
Vastenhouw (10.1016/j.procs.2017.05.009_bib0020) 2005; 47
10.1016/j.procs.2017.05.009_bib0006
10.1016/j.procs.2017.05.009_bib0005
10.1016/j.procs.2017.05.009_bib0016
10.1016/j.procs.2017.05.009_bib0004
10.1016/j.procs.2017.05.009_bib0015
10.1016/j.procs.2017.05.009_bib0014
Liua (10.1016/j.procs.2017.05.009_bib0010) 2013; 55
10.1016/j.procs.2017.05.009_bib0019
10.1016/j.procs.2017.05.009_bib0007
10.1016/j.procs.2017.05.009_bib0018
10.1016/j.procs.2017.05.009_bib0013
10.1016/j.procs.2017.05.009_bib0001
Bento (10.1016/j.procs.2017.05.009_bib0002) 2014; 42
Harper (10.1016/j.procs.2017.05.009_bib0008) 2015; 5
Simm (10.1016/j.procs.2017.05.009_bib0017) 2015
10.1016/j.procs.2017.05.009_bib0022
References_xml – reference: , 2010.
– reference: Tom Vander Aa, Imen Chakroun, and Tom Haber. Distributed bayesian probabilistic matrix factorization. In 2016 IEEE International Conference on Cluster Computing (CLUSTER), pages 346–349, Sept 2016.
– year: 2015
  ident: bib0017
  publication-title: Macau: Scalable bayesian multi-relational factorization with side information using MCMC
– reference: Gaël Guennebaud, Benoît Jacob, et al. Eigen v3.
– reference: MPI Forum. MPI: a message-passing interface standard. Technical report, Knoxville, TN, USA, 1994.
– volume: 5
  start-page: 19:1
  year: December 2015
  end-page: 19:19
  ident: bib0008
  article-title: The movielens datasets: History and context
  publication-title: ACM Trans. Interact. Intell. Syst.
– start-page: 41
  year: 2015
  end-page: 48
  ident: bib0003
  article-title: ExaSHARK: a scalable hybrid array kit for exascale simulation
  publication-title: In Proceedings of the Symposium on High Performance Computing, HPC 2015, part of the 2015 Spring Simulation Multiconference, SpringSim ‘15, Alexandria, VA, USA, April 12-15, 2015
– volume: 55
  start-page: 838
  year: June 2013
  end-page: 850
  ident: bib0010
  article-title: Bayesian probabilistic matrix factorization with social relations and item contents for recommendation
  publication-title: Decision Support Systems
– volume: 55
  start-page: 838
  year: 2013
  end-page: 850
  ident: bib0012
  article-title: Bayesian probabilistic matrix factorization with social relations and item contents for recommendation
  publication-title: Decision Support Systems
– reference: Ruslan Salakhutdinov and Andriy Mnih. Bayesian probabilistic matrix factorization using Markov chain Monte Carlo. In Proceedings of the International Conference on Machine Learning, volume 25, 2008.
– reference: James Reinders. Intel Threading Building Blocks. O’Reilly & Associates, Inc., Sebastopol, CA, USA, first edition, 2007.
– volume: 22
  start-page: 679
  year: 2006
  end-page: 688
  ident: bib0009
  article-title: Another look at measures of forecast accuracy
  publication-title: International Journal of Forecasting
– reference: Yong Guo, Ana Lucia Varbanescu, Alexandru Iosup, Claudio Martella, and Theodore L. Willke. Benchmarking graph-processing platforms: A vision. In Proceedings of the 5th ACM/SPEC International Conference on Performance Engineering, ICPE ‘14, pages 289–292, New York, NY, USA, 2014. ACM.
– volume: 42
  start-page: 30
  year: Aug 2009
  end-page: 37
  ident: bib0011
  article-title: Matrix factorization techniques for recommender systems
  publication-title: Computer
– start-page: 1269
  year: 2012
  end-page: 1277
  ident: bib0021
  article-title: Bayesian group factor analysis
  publication-title: In AISTATS
– reference: Sungjin Ahn, Anoop Korattikara Balan, Nathan Liu, Suju Rajan, and Max Welling. Large-scale distributed bayesian matrix factorization using stochastic gradient MCMC. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2015.
– reference: OpenMP Architecture Review Board. OpenMP application program interface version 3.0, May 2008.
– reference: Yunhong Zhou, Dennis Wilkinson, Robert Schreiber, and Rong Pan. Large-scale parallel collaborative filtering for the Netflix prize. In Proceedings of the 4th International Conference on Algorithmic Aspects in Information and Management, AAIM ‘08, pages 337–348, Berlin, Heidelberg. Springer-Verlag.
– volume: 47
  start-page: 67
  year: 2005
  end-page: 95
  ident: bib0020
  article-title: A two-dimensional data distribution method for parallel sparse matrix-vector multiplication
  publication-title: SIAM Review
– reference: Rainer Gemulla, Erik Nijkamp, Peter J. Haas, and Yannis Sismanis. Large-scale matrix factorization with distributed stochastic gradient descent. In Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ‘11, pages 69–77, New York, NY, USA, 2011. ACM.
– reference: Daniel Grünewald and Christian Simmendinger. The GASPI API specification and its implementation GPI 2.0. In 7th International Conference on PGAS Programming Models, volume 243, 2013.
– reference: Gilbert W. Stewart. Matrix algorithms. Volume 1. , Basic decompositions. SIAM (Society for industrial and applied mathematics), Philadelphia (Pa.), 1998.
– volume: 42
  start-page: 1083
  year: 2014
  end-page: 1090
  ident: bib0002
  article-title: The ChEMBL bioactivity database: an update
  publication-title: Nucleic Acids Res.
– volume: 55
  start-page: 838
  issue: 3
  year: 2013
  ident: 10.1016/j.procs.2017.05.009_bib0012
  article-title: Bayesian probabilistic matrix factorization with social relations and item contents for recommendation
  publication-title: Decision Support Systems
  doi: 10.1016/j.dss.2013.04.002
– start-page: 41
  year: 2015
  ident: 10.1016/j.procs.2017.05.009_bib0003
  article-title: ExaSHARK: a scalable hybrid array kit for exascale simulation
  publication-title: In Proceedings of the Symposium on High Performance Computing, HPC 2015, part of the 2015 Spring Simulation Multiconference, SpringSim ‘15, Alexandria, VA, USA, April 12-15, 2015
– volume: 55
  start-page: 838
  issue: 3
  year: 2013
  ident: 10.1016/j.procs.2017.05.009_bib0010
  article-title: Bayesian probabilistic matrix factorization with social relations and item contents for recommendation
  publication-title: Decision Support Systems
  doi: 10.1016/j.dss.2013.04.002
– ident: 10.1016/j.procs.2017.05.009_bib0014
– ident: 10.1016/j.procs.2017.05.009_bib0015
– ident: 10.1016/j.procs.2017.05.009_bib0018
  doi: 10.1137/1.9781611971408
– volume: 47
  start-page: 67
  issue: 1
  year: 2005
  ident: 10.1016/j.procs.2017.05.009_bib0020
  article-title: A two-dimensional data distribution method for parallel sparse matrix-vector multiplication
  publication-title: SIAM Review
  doi: 10.1137/S0036144502409019
– ident: 10.1016/j.procs.2017.05.009_bib0004
  doi: 10.1145/2020408.2020426
– volume: 22
  start-page: 679
  issue: 4
  year: 2006
  ident: 10.1016/j.procs.2017.05.009_bib0009
  article-title: Another look at measures of forecast accuracy
  publication-title: International Journal of Forecasting
  doi: 10.1016/j.ijforecast.2006.03.001
– ident: 10.1016/j.procs.2017.05.009_bib0013
– volume: 42
  start-page: 1083
  year: 2014
  ident: 10.1016/j.procs.2017.05.009_bib0002
  article-title: The ChEMBL bioactivity database: an update
  publication-title: Nucleic Acids Res.
  doi: 10.1093/nar/gkt1031
– volume: 42
  start-page: 30
  issue: 8
  year: 2009
  ident: 10.1016/j.procs.2017.05.009_bib0011
  article-title: Matrix factorization techniques for recommender systems
  publication-title: Computer
  doi: 10.1109/MC.2009.263
– volume: 5
  start-page: 19:1
  issue: 4
  year: 2015
  ident: 10.1016/j.procs.2017.05.009_bib0008
  article-title: The movielens datasets: History and context
  publication-title: ACM Trans. Interact. Intell. Syst.
  doi: 10.1145/2827872
– ident: 10.1016/j.procs.2017.05.009_bib0007
  doi: 10.1145/2568088.2576761
– ident: 10.1016/j.procs.2017.05.009_bib0001
  doi: 10.1145/2783258.2783373
– ident: 10.1016/j.procs.2017.05.009_bib0022
  doi: 10.1007/978-3-540-68880-8_32
– year: 2015
  ident: 10.1016/j.procs.2017.05.009_bib0017
– ident: 10.1016/j.procs.2017.05.009_bib0019
  doi: 10.1109/CLUSTER.2016.13
– ident: 10.1016/j.procs.2017.05.009_bib0005
– start-page: 1269
  year: 2012
  ident: 10.1016/j.procs.2017.05.009_bib0021
  article-title: Bayesian group factor analysis
  publication-title: In AISTATS
– ident: 10.1016/j.procs.2017.05.009_bib0006
– ident: 10.1016/j.procs.2017.05.009_bib0016
  doi: 10.1145/1390156.1390267
SSID ssj0000388917
Score 2.1753118
Snippet Using the matrix factorization technique in machine learning is very common mainly in areas like rec-ommender systems. Despite its high prediction accuracy and...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 1030
SubjectTerms Collaborative filtering
Machine learning
multi-core
PGAS
Probabilistic matrix factorization algorithm
Title Distributed Bayesian Probabilistic Matrix Factorization
URI https://dx.doi.org/10.1016/j.procs.2017.05.009
Volume 108
WOSCitedRecordID wos000404959000104&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources (ISSN International Center)
  customDbUrl:
  eissn: 1877-0509
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000388917
  issn: 1877-0509
  databaseCode: M~E
  dateStart: 20100101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3PT9swFLag7MBlG2wI2Jhy4NZFipM4do5so-IwEBIF9RbZTiyBIKBSpnLhb-c9_yipqBAg7ZK2Vly3_tzn59f3vo-QXVZrUVMuY1WbMs6lobHKDYsNNzm0KEmtfNvZX350JEaj8tjLW91aOQHetmI6LW_-K9TQBmBj6ewb4J69KTTAcwAdrgA7XF8F_B-kwkUVK3Alf8n7xlZJHo_hd4t5sEjL3D9EXv5pf2C1dnwhZtdLtdUDsHBswjlqPvT9RhngObM1Mf0963kO3ZeYJQmA_bCW5-qpyuxAKr80_L0-zuAKKl3Y61npi7WUgvMYyWPcRrKgLZjXRHQMJKqadTZb_CN6oSF3MYUL3EY0sqpTR7AahptjyD7BcXFYCuaKJXS0TFZSDiclzOR8eAq5IfFNaTWYZx808FDZjL9nYy32VTr-x_Az-egPDtGeA3yNLDXtOvkURDkib6O_EN7BPwr4R3P4Rw7_aA7_r-R0sD_8fRB7dYxYZ7mYxDotlGpqWZcNHAnLRCdZViQqMWmhaWZyY5RhJi2FzlnZ6ELBA2OGZjQteFPobIP02uu22SSRqkUhuZAZkvcoKQVjXOLBOdXgv9T5FknDPFTaU8ejgsllFXIELyo7eRVOXpWwCiZvi_ycdbpxzCkv316ECa78mnZOXQVL4qWO2-_t-I2s4isXT_tOepPxXbNDPuh_k_Pb8Q-7dB4BlK9_2Q
linkProvider ISSN International Centre
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Distributed+Bayesian+Probabilistic+Matrix+Factorization&rft.jtitle=Procedia+computer+science&rft.au=Vander+Aa%2C+Tom&rft.au=Chakroun%2C+Imen&rft.au=Haber%2C+Tom&rft.date=2017&rft.pub=Elsevier+B.V&rft.issn=1877-0509&rft.eissn=1877-0509&rft.volume=108&rft.spage=1030&rft.epage=1039&rft_id=info:doi/10.1016%2Fj.procs.2017.05.009&rft.externalDocID=S187705091730501X
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1877-0509&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1877-0509&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1877-0509&client=summon