HGNN+: General Hypergraph Neural Networks

Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on pattern analysis and machine intelligence Vol. 45; no. 3; pp. 3181 - 3199
Main Authors: Gao, Yue, Feng, Yifan, Ji, Shuyi, Ji, Rongrong
Format: Journal Article
Language:English
Published: United States IEEE 01.03.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have recently been proposed to address the problem of multi-modal/multi-type data correlation by directly concatenating the hypergraphs constructed from each single individual modality/type, which is difficult to learn an adaptive weight for each modality/type. In this paper, we extend the original conference version HGNN, and introduce a general high-order multi-modal/multi-type data correlation modeling framework called HGNN<inline-formula><tex-math notation="LaTeX">^+</tex-math> <mml:math><mml:msup><mml:mrow/><mml:mo>+</mml:mo></mml:msup></mml:math><inline-graphic xlink:href="feng-ieq1-3182052.gif"/> </inline-formula> to learn an optimal representation in a single hypergraph based framework. It is achieved by bridging multi-modal/multi-type data and hyperedge with hyperedge groups. Specifically, in our method, hyperedge groups are first constructed to represent latent high-order correlations in each specific modality/type with explicit or implicit graph structures. An adaptive hyperedge group fusion strategy is then used to effectively fuse the correlations from different modalities/types in a unified hypergraph. After that a new hypergraph convolution scheme performed in spatial domain is used to learn a general data representation for various tasks. We have evaluated this framework on several popular datasets and compared it with recent state-of-the-art methods. The comprehensive evaluations indicate that the proposed HGNN<inline-formula><tex-math notation="LaTeX">^+</tex-math> <mml:math><mml:msup><mml:mrow/><mml:mo>+</mml:mo></mml:msup></mml:math><inline-graphic xlink:href="feng-ieq2-3182052.gif"/> </inline-formula> framework can consistently outperform existing methods with a significant margin, especially when modeling implicit data correlations. We also release a toolbox called THU-DeepHypergraph for the proposed framework, which can be used for various of applications, such as data classification, retrieval and recommendation.
AbstractList Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have recently been proposed to address the problem of multi-modal/multi-type data correlation by directly concatenating the hypergraphs constructed from each single individual modality/type, which is difficult to learn an adaptive weight for each modality/type. In this paper, we extend the original conference version HGNN, and introduce a general high-order multi-modal/multi-type data correlation modeling framework called HGNN<inline-formula><tex-math notation="LaTeX">^+</tex-math> <mml:math><mml:msup><mml:mrow/><mml:mo>+</mml:mo></mml:msup></mml:math><inline-graphic xlink:href="feng-ieq1-3182052.gif"/> </inline-formula> to learn an optimal representation in a single hypergraph based framework. It is achieved by bridging multi-modal/multi-type data and hyperedge with hyperedge groups. Specifically, in our method, hyperedge groups are first constructed to represent latent high-order correlations in each specific modality/type with explicit or implicit graph structures. An adaptive hyperedge group fusion strategy is then used to effectively fuse the correlations from different modalities/types in a unified hypergraph. After that a new hypergraph convolution scheme performed in spatial domain is used to learn a general data representation for various tasks. We have evaluated this framework on several popular datasets and compared it with recent state-of-the-art methods. The comprehensive evaluations indicate that the proposed HGNN<inline-formula><tex-math notation="LaTeX">^+</tex-math> <mml:math><mml:msup><mml:mrow/><mml:mo>+</mml:mo></mml:msup></mml:math><inline-graphic xlink:href="feng-ieq2-3182052.gif"/> </inline-formula> framework can consistently outperform existing methods with a significant margin, especially when modeling implicit data correlations. We also release a toolbox called THU-DeepHypergraph for the proposed framework, which can be used for various of applications, such as data classification, retrieval and recommendation.
Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have recently been proposed to address the problem of multi-modal/multi-type data correlation by directly concatenating the hypergraphs constructed from each single individual modality/type, which is difficult to learn an adaptive weight for each modality/type. In this paper, we extend the original conference version HGNN, and introduce a general high-order multi-modal/multi-type data correlation modeling framework called HGNN to learn an optimal representation in a single hypergraph based framework. It is achieved by bridging multi-modal/multi-type data and hyperedge with hyperedge groups. Specifically, in our method, hyperedge groups are first constructed to represent latent high-order correlations in each specific modality/type with explicit or implicit graph structures. An adaptive hyperedge group fusion strategy is then used to effectively fuse the correlations from different modalities/types in a unified hypergraph. After that a new hypergraph convolution scheme performed in spatial domain is used to learn a general data representation for various tasks. We have evaluated this framework on several popular datasets and compared it with recent state-of-the-art methods. The comprehensive evaluations indicate that the proposed HGNN framework can consistently outperform existing methods with a significant margin, especially when modeling implicit data correlations. We also release a toolbox called THU-DeepHypergraph for the proposed framework, which can be used for various of applications, such as data classification, retrieval and recommendation.
Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have recently been proposed to address the problem of multi-modal/multi-type data correlation by directly concatenating the hypergraphs constructed from each single individual modality/type, which is difficult to learn an adaptive weight for each modality/type. In this paper, we extend the original conference version HGNN, and introduce a general high-order multi-modal/multi-type data correlation modeling framework called HGNN[Formula Omitted] to learn an optimal representation in a single hypergraph based framework. It is achieved by bridging multi-modal/multi-type data and hyperedge with hyperedge groups. Specifically, in our method, hyperedge groups are first constructed to represent latent high-order correlations in each specific modality/type with explicit or implicit graph structures. An adaptive hyperedge group fusion strategy is then used to effectively fuse the correlations from different modalities/types in a unified hypergraph. After that a new hypergraph convolution scheme performed in spatial domain is used to learn a general data representation for various tasks. We have evaluated this framework on several popular datasets and compared it with recent state-of-the-art methods. The comprehensive evaluations indicate that the proposed HGNN[Formula Omitted] framework can consistently outperform existing methods with a significant margin, especially when modeling implicit data correlations. We also release a toolbox called THU-DeepHypergraph for the proposed framework, which can be used for various of applications, such as data classification, retrieval and recommendation.
Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have recently been proposed to address the problem of multi-modal/multi-type data correlation by directly concatenating the hypergraphs constructed from each single individual modality/type, which is difficult to learn an adaptive weight for each modality/type. In this paper, we extend the original conference version HGNN, and introduce a general high-order multi-modal/multi-type data correlation modeling framework called HGNN + to learn an optimal representation in a single hypergraph based framework. It is achieved by bridging multi-modal/multi-type data and hyperedge with hyperedge groups. Specifically, in our method, hyperedge groups are first constructed to represent latent high-order correlations in each specific modality/type with explicit or implicit graph structures. An adaptive hyperedge group fusion strategy is then used to effectively fuse the correlations from different modalities/types in a unified hypergraph. After that a new hypergraph convolution scheme performed in spatial domain is used to learn a general data representation for various tasks. We have evaluated this framework on several popular datasets and compared it with recent state-of-the-art methods. The comprehensive evaluations indicate that the proposed HGNN + framework can consistently outperform existing methods with a significant margin, especially when modeling implicit data correlations. We also release a toolbox called THU-DeepHypergraph for the proposed framework, which can be used for various of applications, such as data classification, retrieval and recommendation.Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits their applications in dealing with complex data correlation of multi-modal/multi-type data in practice. A few hypergraph-based methods have recently been proposed to address the problem of multi-modal/multi-type data correlation by directly concatenating the hypergraphs constructed from each single individual modality/type, which is difficult to learn an adaptive weight for each modality/type. In this paper, we extend the original conference version HGNN, and introduce a general high-order multi-modal/multi-type data correlation modeling framework called HGNN + to learn an optimal representation in a single hypergraph based framework. It is achieved by bridging multi-modal/multi-type data and hyperedge with hyperedge groups. Specifically, in our method, hyperedge groups are first constructed to represent latent high-order correlations in each specific modality/type with explicit or implicit graph structures. An adaptive hyperedge group fusion strategy is then used to effectively fuse the correlations from different modalities/types in a unified hypergraph. After that a new hypergraph convolution scheme performed in spatial domain is used to learn a general data representation for various tasks. We have evaluated this framework on several popular datasets and compared it with recent state-of-the-art methods. The comprehensive evaluations indicate that the proposed HGNN + framework can consistently outperform existing methods with a significant margin, especially when modeling implicit data correlations. We also release a toolbox called THU-DeepHypergraph for the proposed framework, which can be used for various of applications, such as data classification, retrieval and recommendation.
Author Feng, Yifan
Ji, Rongrong
Gao, Yue
Ji, Shuyi
Author_xml – sequence: 1
  givenname: Yue
  orcidid: 0000-0002-4971-590X
  surname: Gao
  fullname: Gao, Yue
  email: kevin.gaoy@gmail.com
  organization: BNRist, KLISS, School of Software, BLBCI, THUIBCS, Tsinghua University, Beijing, China
– sequence: 2
  givenname: Yifan
  orcidid: 0000-0003-0878-2986
  surname: Feng
  fullname: Feng, Yifan
  email: evanfeng97@gmail.com
  organization: BNRist, KLISS, School of Software, BLBCI, THUIBCS, Tsinghua University, Beijing, China
– sequence: 3
  givenname: Shuyi
  orcidid: 0000-0003-3795-3545
  surname: Ji
  fullname: Ji, Shuyi
  email: jisy19@mails.tsinghua.edu.cn
  organization: BNRist, KLISS, School of Software, BLBCI, THUIBCS, Tsinghua University, Beijing, China
– sequence: 4
  givenname: Rongrong
  orcidid: 0000-0001-9163-2932
  surname: Ji
  fullname: Ji, Rongrong
  email: rrji@xmu.edu.cn
  organization: Media Analytics and Computing Laboratory, Department of Artificial Intelligence, School of Informatics, Institute of Artificial Intelligence, Fujian Engineering Research Center of Trusted Artificial Intelligence Analysis and Application, Xiamen University, Xiamen, Fujian, China
BackLink https://www.ncbi.nlm.nih.gov/pubmed/35696461$$D View this record in MEDLINE/PubMed
BookMark eNp9kE1Lw0AQhhdRtFb_gIIUvCiSujv77a2ItoJWD3peNulEo2lSdxPEf29qqwcPzmVgeN6Z4dklm1VdISEHjA4Zo_b88WF0dzMECjDkzACVsEF6wBRNLFjYJD3KFCTGgNkhuzG-UsqEpHyb7HCprBKK9cjpZDydnl0Mxlhh8OVg8rnA8Bz84mUwxXY5mWLzUYe3uEe2cl9G3F_3Pnm6vnq8nCS39-Oby9FtknHJmoSLTCpmUSs6M1awzMpU6BSlEWKmvOK5TRFMamimc5NCTjPuU8G1mYHVXvA-OVntXYT6vcXYuHkRMyxLX2HdRgdKKymZVdChx3_Q17oNVfedA625kJJ31SdHa6pN5zhzi1DMffh0PxI6AFZAFuoYA-a_CKNuadp9m3ZL025tuguZP6GsaHxT1FUTfFH-Hz1cRQtE_L1ltZUgGf8CxHOHVg
CODEN ITPIDJ
CitedBy_id crossref_primary_10_1016_j_patcog_2025_111864
crossref_primary_10_1109_TGRS_2025_3567746
crossref_primary_10_26599_BDMA_2024_9020094
crossref_primary_10_1007_s10489_024_05939_4
crossref_primary_10_1016_j_knosys_2024_111903
crossref_primary_10_1155_ijdm_7220522
crossref_primary_10_1109_TGRS_2025_3565203
crossref_primary_10_1109_TIFS_2025_3580218
crossref_primary_10_1007_s10723_025_09807_4
crossref_primary_10_1109_TNSE_2025_3547349
crossref_primary_10_1109_TNSRE_2025_3595655
crossref_primary_10_1145_3745020
crossref_primary_10_3390_s24237655
crossref_primary_10_1016_j_sigpro_2024_109797
crossref_primary_10_1109_TBDATA_2025_3527216
crossref_primary_10_1007_s10723_024_09767_1
crossref_primary_10_1007_s42979_024_03453_5
crossref_primary_10_1016_j_mineng_2025_109764
crossref_primary_10_1016_j_neucom_2024_128539
crossref_primary_10_1109_TPAMI_2024_3353199
crossref_primary_10_3390_electronics13071306
crossref_primary_10_1038_s41598_025_01856_9
crossref_primary_10_1016_j_patcog_2025_112397
crossref_primary_10_1007_s44295_025_00070_7
crossref_primary_10_1109_JBHI_2024_3434394
crossref_primary_10_1360_SSI_2024_0282
crossref_primary_10_1007_s13042_024_02414_x
crossref_primary_10_1016_j_inffus_2023_101939
crossref_primary_10_1016_j_trc_2025_105271
crossref_primary_10_1007_s11263_024_02298_y
crossref_primary_10_1016_j_neunet_2025_107451
crossref_primary_10_1093_bib_bbae355
crossref_primary_10_1109_JSTARS_2024_3360431
crossref_primary_10_1016_j_ins_2023_119412
crossref_primary_10_1007_s40305_025_00630_y
crossref_primary_10_1109_TNNLS_2025_3542176
crossref_primary_10_1109_TBDATA_2025_3533908
crossref_primary_10_1109_JSEN_2024_3492041
crossref_primary_10_1016_j_neucom_2023_126992
crossref_primary_10_1109_TKDE_2024_3380643
crossref_primary_10_1109_TBDATA_2024_3442549
crossref_primary_10_1109_TR_2024_3393415
crossref_primary_10_1016_j_ins_2025_121960
crossref_primary_10_1007_s11227_024_06003_1
crossref_primary_10_1109_TAI_2024_3524984
crossref_primary_10_1057_s41599_025_05180_5
crossref_primary_10_1109_TNNLS_2024_3422265
crossref_primary_10_1016_j_knosys_2024_112177
crossref_primary_10_1145_3719013
crossref_primary_10_1109_ACCESS_2024_3398367
crossref_primary_10_3390_electronics14091902
crossref_primary_10_1016_j_patcog_2025_111921
crossref_primary_10_1109_TGRS_2025_3543556
crossref_primary_10_1016_j_amc_2025_129698
crossref_primary_10_1109_TGRS_2025_3534288
crossref_primary_10_1016_j_patcog_2025_111926
crossref_primary_10_1145_3762665
crossref_primary_10_1088_2632_2153_adf375
crossref_primary_10_1016_j_neucom_2024_127989
crossref_primary_10_1109_TMM_2024_3521738
crossref_primary_10_1016_j_energy_2025_135170
crossref_primary_10_1140_epjb_s10051_024_00791_4
crossref_primary_10_3390_e26030239
crossref_primary_10_1016_j_asoc_2025_113644
crossref_primary_10_1016_j_asoc_2025_113645
crossref_primary_10_1109_TCSVT_2024_3359681
crossref_primary_10_1016_j_ipm_2023_103518
crossref_primary_10_1016_j_jmapro_2025_08_043
crossref_primary_10_1109_TMM_2025_3535298
crossref_primary_10_1093_bib_bbae658
crossref_primary_10_1371_journal_pone_0310189
crossref_primary_10_1016_j_ipm_2024_103877
crossref_primary_10_3390_biology12091203
crossref_primary_10_1016_j_patcog_2025_111995
crossref_primary_10_1109_TNNLS_2024_3368111
crossref_primary_10_3390_a16030126
crossref_primary_10_1109_TIP_2025_3592551
crossref_primary_10_1016_j_patcog_2024_110364
crossref_primary_10_1109_TII_2024_3393137
crossref_primary_10_1109_TNNLS_2024_3456593
crossref_primary_10_1038_s41598_025_05891_4
crossref_primary_10_1109_TCOMM_2024_3511952
crossref_primary_10_3390_rs16183533
crossref_primary_10_1016_j_asoc_2025_112721
crossref_primary_10_1109_TPAMI_2024_3524377
crossref_primary_10_1109_TRO_2025_3600144
crossref_primary_10_1016_j_cmpb_2024_108237
crossref_primary_10_1007_s11263_025_02438_y
crossref_primary_10_1007_s10489_025_06348_x
crossref_primary_10_1016_j_rineng_2025_104443
crossref_primary_10_1109_TGRS_2025_3598612
crossref_primary_10_3389_fmed_2024_1496573
crossref_primary_10_3390_app142210646
crossref_primary_10_1016_j_physa_2025_130725
crossref_primary_10_1109_TAI_2024_3450658
crossref_primary_10_1016_j_patcog_2025_111544
crossref_primary_10_1109_TMM_2025_3542958
crossref_primary_10_1109_TBDATA_2024_3453757
crossref_primary_10_1109_TIM_2025_3600734
crossref_primary_10_1109_TMM_2024_3521678
crossref_primary_10_1109_TPAMI_2025_3557391
crossref_primary_10_1021_acs_jcim_5c01164
crossref_primary_10_1109_TKDE_2024_3524627
crossref_primary_10_3390_app14093832
crossref_primary_10_1088_1674_1056_ad20db
crossref_primary_10_1016_j_cities_2025_106334
crossref_primary_10_1016_j_neucom_2025_129835
crossref_primary_10_1145_3733234
crossref_primary_10_1016_j_media_2025_103700
crossref_primary_10_1038_s41467_025_60695_4
crossref_primary_10_1007_s10489_025_06491_5
crossref_primary_10_1109_TMC_2025_3556243
crossref_primary_10_1109_TFUZZ_2025_3588125
crossref_primary_10_3390_brainsci14080738
crossref_primary_10_1007_s10115_024_02259_4
crossref_primary_10_1007_s00371_024_03541_w
crossref_primary_10_1145_3664651
crossref_primary_10_1007_s10489_025_06685_x
crossref_primary_10_1145_3735135
crossref_primary_10_1007_s10462_025_11199_6
crossref_primary_10_1016_j_media_2025_103661
crossref_primary_10_1016_j_media_2024_103144
crossref_primary_10_1093_bib_bbae718
crossref_primary_10_1109_TCSVT_2025_3535930
crossref_primary_10_1093_bib_bbae274
crossref_primary_10_1016_j_bspc_2024_107075
crossref_primary_10_1109_TPAMI_2024_3459932
crossref_primary_10_3389_fmed_2025_1654199
crossref_primary_10_14778_3749646_3749692
crossref_primary_10_1287_ijoc_2024_0660
crossref_primary_10_1016_j_eswa_2023_119842
crossref_primary_10_4018_IJDWM_349975
crossref_primary_10_1007_s10618_023_00956_2
crossref_primary_10_1109_TGRS_2024_3440271
crossref_primary_10_1109_TBDATA_2023_3278988
crossref_primary_10_1007_s10115_025_02476_5
crossref_primary_10_1007_s41109_023_00568_1
crossref_primary_10_1109_TKDE_2025_3539769
crossref_primary_10_1007_s00521_024_10365_1
crossref_primary_10_1109_TGRS_2025_3542422
crossref_primary_10_1109_TKDE_2025_3570098
crossref_primary_10_1109_TVCG_2024_3381152
crossref_primary_10_1109_TAES_2025_3560256
crossref_primary_10_1016_j_ins_2025_122484
crossref_primary_10_1109_TKDE_2024_3435861
crossref_primary_10_1016_j_media_2024_103368
crossref_primary_10_1016_j_scs_2024_105366
crossref_primary_10_1109_TGRS_2023_3345159
crossref_primary_10_1016_j_inffus_2024_102769
crossref_primary_10_1016_j_ipm_2024_103958
crossref_primary_10_1016_j_neucom_2024_128761
crossref_primary_10_1038_s41467_025_61149_7
crossref_primary_10_1145_3605776
crossref_primary_10_1016_j_physa_2024_129891
crossref_primary_10_1109_JSTARS_2024_3403863
crossref_primary_10_1145_3744918
crossref_primary_10_1109_TIM_2025_3566845
crossref_primary_10_1016_j_patcog_2025_112125
crossref_primary_10_1109_ACCESS_2025_3592104
crossref_primary_10_1109_TIP_2025_3570604
crossref_primary_10_1109_TPAMI_2025_3585179
crossref_primary_10_1016_j_inffus_2025_103016
crossref_primary_10_1016_j_eneco_2025_108895
crossref_primary_10_1109_TMC_2024_3515154
crossref_primary_10_1007_s40747_025_02016_2
crossref_primary_10_1007_s11227_025_07835_1
crossref_primary_10_1145_3653015
crossref_primary_10_1016_j_bspc_2025_108049
crossref_primary_10_1016_j_knosys_2025_114435
crossref_primary_10_3390_electronics13224435
crossref_primary_10_1016_j_eswa_2025_127242
crossref_primary_10_3389_fdgth_2024_1503831
crossref_primary_10_1016_j_neunet_2024_106562
crossref_primary_10_1109_TSIPN_2025_3599780
Cites_doi 10.12732/ijpam.v98i3.2
10.1109/ICCV.2015.114
10.1103/PhysRevE.101.022308
10.1016/j.cviu.2013.10.012
10.1145/2043932.2044016
10.1109/CVPR.2018.00035
10.1609/aaai.v33i01.33014602
10.1016/j.acha.2010.04.005
10.1145/3219819.3219829
10.1145/1143844.1143847
10.1117/12.2543089
10.1016/j.patcog.2020.107637
10.1145/3331184.3331267
10.1609/aimag.v29i3.2157
10.1109/TIP.2018.2837219
10.1109/MSP.2012.2235192
10.1109/43.784130
10.7551/mitpress/7503.003.0205
10.1016/0166-218X(93)90045-P
10.1109/BHI.2019.8834572
10.24963/ijcai.2019/366
10.1609/aaai.v33i01.33013558
10.1111/1467-8659.00669
10.1093/comnet/cnab014
10.1109/CVPR.2016.90
10.1145/3366423.3380152
10.1109/CVPR.2015.7298801
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
DBID 97E
RIA
RIE
AAYXX
CITATION
NPM
7SC
7SP
8FD
JQ2
L7M
L~C
L~D
7X8
DOI 10.1109/TPAMI.2022.3182052
DatabaseName IEEE Xplore (IEEE)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
PubMed
Computer and Information Systems Abstracts
Electronics & Communications Abstracts
Technology Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
Technology Research Database
Computer and Information Systems Abstracts – Academic
Electronics & Communications Abstracts
ProQuest Computer Science Collection
Computer and Information Systems Abstracts
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitleList
PubMed
Technology Research Database
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
EISSN 2160-9292
1939-3539
EndPage 3199
ExternalDocumentID 35696461
10_1109_TPAMI_2022_3182052
9795251
Genre orig-research
Journal Article
GrantInformation_xml – fundername: Open Research Projects of Zhejiang Lab
  grantid: 2021KG0AB05
– fundername: National Science Fund for Distinguished Young Scholars
  grantid: 62025603
  funderid: 10.13039/501100014219
– fundername: National Natural Science Foundation of China; National Natural Science Funds of China
  grantid: 62088102; 62021002
  funderid: 10.13039/501100001809
GroupedDBID ---
-DZ
-~X
.DC
0R~
29I
4.4
53G
5GY
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACGFS
ACIWK
ACNCT
AENEX
AGQYO
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
E.L
EBS
EJD
F5P
HZ~
IEDLZ
IFIPE
IPLJI
JAVBF
LAI
M43
MS~
O9-
OCL
P2P
PQQKQ
RIA
RIE
RNS
RXW
TAE
TN5
UHB
~02
AAYXX
CITATION
5VS
9M8
AAYOK
ABFSI
ADRHT
AETIX
AGSQL
AI.
AIBXA
ALLEH
FA8
H~9
IBMZZ
ICLAB
IFJZH
NPM
PKN
RIC
RIG
RNI
RZB
VH1
XJT
Z5M
7SC
7SP
8FD
JQ2
L7M
L~C
L~D
7X8
ID FETCH-LOGICAL-c351t-34c5619e760d8941c95b47be5844d6a63f9be28b80c7f8b2f0c3ab4378d297a43
IEDL.DBID RIE
ISICitedReferencesCount 257
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000966221400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0162-8828
1939-3539
IngestDate Wed Oct 01 12:34:41 EDT 2025
Mon Jun 30 07:09:06 EDT 2025
Wed Feb 19 02:24:39 EST 2025
Tue Nov 18 21:01:02 EST 2025
Sat Nov 29 02:58:20 EST 2025
Wed Aug 27 02:06:04 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 3
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c351t-34c5619e760d8941c95b47be5844d6a63f9be28b80c7f8b2f0c3ab4378d297a43
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ORCID 0000-0002-4971-590X
0000-0003-0878-2986
0000-0003-3795-3545
0000-0001-9163-2932
PMID 35696461
PQID 2773455333
PQPubID 85458
PageCount 19
ParticipantIDs pubmed_primary_35696461
proquest_miscellaneous_2676551962
crossref_primary_10_1109_TPAMI_2022_3182052
crossref_citationtrail_10_1109_TPAMI_2022_3182052
ieee_primary_9795251
proquest_journals_2773455333
PublicationCentury 2000
PublicationDate 2023-03-01
PublicationDateYYYYMMDD 2023-03-01
PublicationDate_xml – month: 03
  year: 2023
  text: 2023-03-01
  day: 01
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
– name: New York
PublicationTitle IEEE transactions on pattern analysis and machine intelligence
PublicationTitleAbbrev TPAMI
PublicationTitleAlternate IEEE Trans Pattern Anal Mach Intell
PublicationYear 2023
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References ref13
ref11
ref10
Defferrard (ref5)
ref19
ref18
Hamilton (ref1)
Xu (ref17)
You (ref15)
Shchur (ref29)
ref45
Hein (ref48)
Chien (ref46)
ref42
ref41
ref43
Veličković (ref3)
ref7
ref9
ref6
ref40
Yang (ref44) 2020
Aldous (ref38) 2002
Yadati (ref8)
ref34
ref37
ref36
ref31
ref30
ref33
ref32
Kipf (ref2)
Defferrard (ref49)
Namata (ref26)
Atwood (ref16)
Kelly (ref39) 2011
ref23
ref25
ref20
ref22
ref21
Gilmer (ref4)
ref28
ref27
Chitra (ref35)
Shervashidze (ref24) 2011; 9
Henaff (ref14) 2015
Li (ref47)
Bruna (ref12)
References_xml – ident: ref10
  doi: 10.12732/ijpam.v98i3.2
– ident: ref42
  doi: 10.1109/ICCV.2015.114
– ident: ref36
  doi: 10.1103/PhysRevE.101.022308
– ident: ref34
  doi: 10.1016/j.cviu.2013.10.012
– ident: ref32
  doi: 10.1145/2043932.2044016
– start-page: 1172
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref35
  article-title: Random walks on hypergraphs with edge-dependent vertex weights
– start-page: 2466
  volume-title: Proc. 22nd Int. Conf. Artif. Intell. Statist.
  ident: ref46
  article-title: $HS^{2}$HS2: Active learning over hypergraphs with pointwise and pairwise queries
– volume-title: Proc. Int. Conf. Learn. Representations
  ident: ref12
  article-title: Spectral networks and locally connected networks on graphs
– year: 2015
  ident: ref14
  article-title: Deep convolutional networks on graph-structured data
– year: 2002
  ident: ref38
  article-title: Reversible markov chains and random walks on graphs
– volume-title: Proc. Workshop RRL
  ident: ref29
  article-title: Pitfalls of graph neural network evaluation
– ident: ref43
  doi: 10.1109/CVPR.2018.00035
– ident: ref28
  doi: 10.1609/aaai.v33i01.33014602
– year: 2020
  ident: ref44
  article-title: Hypergraph learning with line expansion
– start-page: 1263
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref4
  article-title: Neural message passing for quantum chemistry
– volume-title: Proc. Int. Conf. Neural Inf. Process. Syst.
  ident: ref48
  article-title: The total variation on hypergraphs-learning on hypergraphs revisited
– ident: ref13
  doi: 10.1016/j.acha.2010.04.005
– ident: ref37
  doi: 10.1145/3219819.3219829
– start-page: 1024
  volume-title: Proc. Int. Conf. Neural Inf. Process. Syst.
  ident: ref1
  article-title: Inductive representation learning on large graphs
– start-page: 3844
  volume-title: Proc. Int. Conf. Neural Inf. Process. Syst.
  ident: ref5
  article-title: Convolutional neural networks on graphs with fast localized spectral filtering
– ident: ref33
  doi: 10.1145/1143844.1143847
– ident: ref22
  doi: 10.1117/12.2543089
– ident: ref7
  doi: 10.1016/j.patcog.2020.107637
– ident: ref19
  doi: 10.1145/3331184.3331267
– ident: ref25
  doi: 10.1609/aimag.v29i3.2157
– start-page: 1993
  volume-title: Proc. Int. Conf. Neural Inf. Process. Syst.
  ident: ref16
  article-title: Diffusion-convolutional neural networks
– ident: ref21
  doi: 10.1109/TIP.2018.2837219
– volume-title: Proc. Int. Conf. Neural Inf. Process. Syst.
  ident: ref49
  article-title: Convolutional neural networks on graphs with fast localized spectral filtering
– start-page: 3014
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref47
  article-title: Submodular hypergraphs: P-Laplacians, cheeger inequalities and spectral clustering
– ident: ref11
  doi: 10.1109/MSP.2012.2235192
– ident: ref45
  doi: 10.1109/43.784130
– ident: ref20
  doi: 10.7551/mitpress/7503.003.0205
– ident: ref23
  doi: 10.1016/0166-218X(93)90045-P
– ident: ref18
  doi: 10.1109/BHI.2019.8834572
– ident: ref9
  doi: 10.24963/ijcai.2019/366
– volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref3
  article-title: Graph attention networks
– volume-title: Proc. Int. Conf. Learn. Representations
  ident: ref17
  article-title: How powerful are graph neural networks?
– volume: 9
  year: 2011
  ident: ref24
  article-title: Weisfeiler-lehman graph kernels
  publication-title: J. Mach. Learn. Res.
– start-page: 7134
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref15
  article-title: Position-aware graph neural networks
– ident: ref6
  doi: 10.1609/aaai.v33i01.33013558
– volume-title: Proc. Int. Conf. Learn. Representations
  ident: ref2
  article-title: Semi-supervised classification with graph convolutional networks
– ident: ref41
  doi: 10.1111/1467-8659.00669
– ident: ref27
  doi: 10.1093/comnet/cnab014
– volume-title: Proc. 10th Int. Workshop Mining Learn. Graphs
  ident: ref26
  article-title: Query-driven active surveying for collective classification
– volume-title: Reversibility and Stochastic Networks[M]
  year: 2011
  ident: ref39
– start-page: 1511
  volume-title: Proc. Conf. Assoc. Advance. Artif. Intell.
  ident: ref8
  article-title: HyperGCN: A new method for training graph convolutional networks on hypergraphs
– ident: ref30
  doi: 10.1109/CVPR.2016.90
– ident: ref31
  doi: 10.1145/3366423.3380152
– ident: ref40
  doi: 10.1109/CVPR.2015.7298801
SSID ssj0014503
Score 2.741128
Snippet Graph Neural Networks have attracted increasing attention in recent years. However, existing GNN frameworks are deployed based upon simple graphs, which limits...
SourceID proquest
pubmed
crossref
ieee
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 3181
SubjectTerms classification
Convolution
Correlation
Data correlation
Data models
Evaluation
Graph neural networks
Graph theory
Hypergraph
hypergraph convolution
Mathematical models
Modelling
Neural networks
Representation learning
Representations
Smart structures
Social networking (online)
Task analysis
Title HGNN+: General Hypergraph Neural Networks
URI https://ieeexplore.ieee.org/document/9795251
https://www.ncbi.nlm.nih.gov/pubmed/35696461
https://www.proquest.com/docview/2773455333
https://www.proquest.com/docview/2676551962
Volume 45
WOSCitedRecordID wos000966221400001&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: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 2160-9292
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014503
  issn: 0162-8828
  databaseCode: RIE
  dateStart: 19790101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB6VikM5tNDSslCqIPVSlVDH7-FWIcoiQdRDQXuLYseRKlW7aB_9_Yydh4oElbhFieNEnpn4-2LPNwCnngvlNXM5MzXm0jYhd0XNciODRIHSySRg-vObKUs7m-H1Frwfc2FCCGnzWfgQD9NafrPwm_ir7AINKh7zpZ8YY7pcrXHFQKpUBZkQDEU40YghQYbhxc315fevRAU5J4ZKM56KJWyE0qilLv6Yj1KBlX9jzTTnXO3939s-h90eW2aXnTO8gK0w34e9oW5D1ofxPjx7IEJ4AGfTL2V5_jHrBaizKTHTZdKxzqJyB50pu63iq5fw4-rzzadp3hdQyL1QxToX0hM8wmA0ayzKwqNy0rhAoEM2utaiRRe4dZZ501rHW-ZF7aQwtuFoaikOYXu-mIdXkFnliVgwx03rJRbKGdYi946-AZYYlZpAMQxj5Xt18Vjk4q5KLINhlaxQRStUvRUmcD7e86vT1ni09UEc47FlP7wTOB6sVfXht6q4MUIqQrJiAu_GyxQ4cTWknofFhtpoowkuoqaejzorj30PzvH67898Azux6ny3Fe0YttfLTXgLT_39-na1PCHvnNmT5J2_Aasx2Vk
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB5VLRJwaKHlsaVAkLigEurY4xe3ClG2Yhv1sKDeothxJCS0i_bB72fsPFQkQOIWJRMn8nji74s98wG89lxIr5jLma5tjqYJuStqlmsMaIVFh6mA6deZLktzc2Ovd-DtmAsTQkibz8K7eJjW8pul38ZfZWdWW8ljvvSeRORFl601rhmgTDrIhGEoxolIDCkyzJ7Nr8-vLokMck4cleY8GUVshFRWoSp-m5GSxMrf0WaadS4O_u99H8B-jy6z8244PISdsDiEg0G5IesD-RDu3ypDeARvpp_K8vR91pegzqbETVepknUWa3fQmbLbLL5-BF8uPs4_TPNeQiH3QhabXKAngGSDVqwxFgtvpUPtAsEObFStRGtd4MYZ5nVrHG-ZF7VDoU3Dra5RPIbdxXIRnkJmpCdqwRzXrUdbSKdZa7l39BUwxKnkBIqhGyvf1xePMhffq8QzmK2SF6rohar3wgROx3t-dNU1_ml9FPt4tOy7dwIng7eqPgDXFddaoCQsKybwarxMoRPXQ-pFWG7JRmlFgNEqavlJ5-Wx7WFwHP_5mS_h7nR-Natml-XnZ3AvatB3G9NOYHez2obncMf_3Hxbr16kMfoLB13buA
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=HGNN%2B%3A+General+Hypergraph+Neural+Networks&rft.jtitle=IEEE+transactions+on+pattern+analysis+and+machine+intelligence&rft.au=Gao%2C+Yue&rft.au=Feng%2C+Yifan&rft.au=Ji%2C+Shuyi&rft.au=Ji%2C+Rongrong&rft.date=2023-03-01&rft.pub=The+Institute+of+Electrical+and+Electronics+Engineers%2C+Inc.+%28IEEE%29&rft.issn=0162-8828&rft.eissn=1939-3539&rft.volume=45&rft.issue=3&rft.spage=3181&rft_id=info:doi/10.1109%2FTPAMI.2022.3182052&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0162-8828&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0162-8828&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0162-8828&client=summon