{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,7]],"date-time":"2026-02-07T19:11:35Z","timestamp":1770491495549,"version":"3.49.0"},"reference-count":34,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2020,11,17]],"date-time":"2020-11-17T00:00:00Z","timestamp":1605571200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2020,11,17]],"date-time":"2020-11-17T00:00:00Z","timestamp":1605571200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"DEEPer project"},{"DOI":"10.13039\/501100003500","name":"Universit\u00e0 degli Studi di Padova","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100003500","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Comput &amp; Applic"],"published-print":{"date-parts":[[2022,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Graph property prediction is becoming more and more popular due to the increasing availability of scientific and social data naturally represented in a graph form. Because of that, many researchers are focusing on the development of improved graph neural network models. One of the main components of a graph neural network is the aggregation operator, needed to generate a graph-level representation from a set of node-level embeddings. The aggregation operator is critical since it should, in principle, provide a representation of the graph that is isomorphism invariant, i.e. the graph representation should be a function of graph nodes treated as a set. DeepSets (in: Advances in neural information processing systems, pp 3391\u20133401, 2017) provides a framework to construct a set-aggregation operator with universal approximation properties. In this paper, we propose a DeepSets aggregation operator, based on Self-Organizing Maps (SOM), to transform a set of node-level representations into a single graph-level one. The adoption of SOMs allows to compute node representations that embed the information about their mutual similarity. Experimental results on several real-world datasets show that our proposed approach achieves improved predictive performance compared to the commonly adopted sum aggregation and many state-of-the-art graph neural network architectures in the literature.<\/jats:p>","DOI":"10.1007\/s00521-020-05484-4","type":"journal-article","created":{"date-parts":[[2020,11,17]],"date-time":"2020-11-17T17:19:42Z","timestamp":1605633582000},"page":"5-24","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["SOM-based aggregation for graph convolutional neural networks"],"prefix":"10.1007","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3023-3046","authenticated-orcid":false,"given":"Luca","family":"Pasa","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nicol\u00f2","family":"Navarin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alessandro","family":"Sperduti","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2020,11,17]]},"reference":[{"key":"5484_CR1","unstructured":"Atwood J, Towsley D (2016) Diffusion-convolutional neural networks. In: Advances in neural information processing systems, pp 1993\u20132001. arXiv:1511.02136"},{"issue":"1","key":"5484_CR2","doi-asserted-by":"publisher","first-page":"215","DOI":"10.1162\/089976698300017953","volume":"10","author":"CM Bishop","year":"1998","unstructured":"Bishop CM, Svens\u00e9n M, Williams CK (1998) GTM: The generative topographic mapping. Neural Comput 10(1):215\u2013234","journal-title":"Neural Comput"},{"issue":"1","key":"5484_CR3","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1093\/bioinformatics\/bti1007","volume":"21","author":"KM Borgwardt","year":"2005","unstructured":"Borgwardt KM, Ong CS, Sch\u00f6nauer S, Vishwanathan S, Smola AJ, Kriegel HP (2005) Protein function prediction via graph kernels. Bioinformatics 21(1):47\u201356","journal-title":"Bioinformatics"},{"key":"5484_CR4","unstructured":"Chen T, Bian S, Sun Y (2019) Are powerful graph neural nets necessary? A dissection on graph classification. arXiv:1905.04579"},{"key":"5484_CR5","unstructured":"Defferrard M, Bresson X., Vandergheynst, P.: Convolutional neural networks on graphs with fast localized spectral filtering. In: Neural information processing systems (NIPS) (2016). arXiv:1606.09375"},{"issue":"4","key":"5484_CR6","doi-asserted-by":"publisher","first-page":"771","DOI":"10.1016\/S0022-2836(03)00628-4","volume":"330","author":"PD Dobson","year":"2003","unstructured":"Dobson PD, Doig AJ (2003) Distinguishing enzyme structures from non-enzymes without alignments. J Mol Biol 330(4):771\u2013783","journal-title":"J Mol Biol"},{"key":"5484_CR7","unstructured":"Duvenaud D, Maclaurin D, Aguilera-Iparraguirre J, G\u00f3mez-Bombarelli R, Hirzel T, Aspuru-Guzik A, Adams RP (2015) Convolutional networks on graphs for learning molecular fingerprints. In: Advances in neural information processing systems. Montreal, Canada, pp 2215\u20132223. arXiv:1509.09292"},{"key":"5484_CR8","unstructured":"Errica, F., Podda, M., Bacciu, D., Micheli, A.: A fair comparison of graph neural networks for graph classification. In: International conference on learning representations (2020). https:\/\/openreview.net\/forum?id=HygDF6NFPB"},{"key":"5484_CR9","unstructured":"Gilmer J, Schoenholz SS, Riley PF, Vinyals O, Dahl GE (2017) Neural message passing for quantum chemistry. In: Proceedings of the 34th international conference on machine learning, pp 1263\u20131272. arXiv:1704.01212"},{"key":"5484_CR10","doi-asserted-by":"publisher","unstructured":"Hagenbuchner M, Tsoi AC, Sperduti A (2001) A supervised self-organizing map for structured data. In: Allinson NM, Yin H, Allinson L, Slack JM (eds) Advances in self-organising maps, WSOM 2001, Lincoln, UK, 13\u201315 June. Springer, pp 21\u201328. https:\/\/doi.org\/10.1007\/978-1-4471-0715-6_4","DOI":"10.1007\/978-1-4471-0715-6_4"},{"key":"5484_CR11","unstructured":"Hamilton W, Ying Z, Leskovec J (2017) Inductive representation learning on large graphs. In: NIPS, pp 1024\u20131034"},{"issue":"1","key":"5484_CR12","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1093\/bioinformatics\/17.1.107","volume":"17","author":"C Helma","year":"2001","unstructured":"Helma C, King RD, Kramer S, Srinivasan A (2001) The predictive toxicology challenge 2000\u20132001. Bioinformatics 17(1):107\u2013108","journal-title":"Bioinformatics"},{"key":"5484_CR13","unstructured":"Kipf TN, Welling M (2016) Variational graph auto-encoders. In: NIPS workshop on Bayesian deep learning . arXiv:1611.07308"},{"key":"5484_CR14","doi-asserted-by":"publisher","unstructured":"Kipf TN, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: ICLR, pp 1\u201314. https:\/\/doi.org\/10.1051\/0004-6361\/201527329. arXiv:1609.02907","DOI":"10.1051\/0004-6361\/201527329"},{"key":"5484_CR15","doi-asserted-by":"publisher","unstructured":"Kipf TN, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: ICLR. https:\/\/doi.org\/10.1051\/0004-6361\/201527329. arxiv:1609.02907","DOI":"10.1051\/0004-6361\/201527329"},{"issue":"9","key":"5484_CR16","doi-asserted-by":"publisher","first-page":"1464","DOI":"10.1109\/5.58325","volume":"78","author":"T Kohonen","year":"1990","unstructured":"Kohonen T (1990) The self-organizing map. Proc IEEE 78(9):1464\u20131480","journal-title":"Proc IEEE"},{"key":"5484_CR17","doi-asserted-by":"publisher","unstructured":"Li, Y., Tarlow, D., Brockschmidt, M., Zemel, R.: Gated graph sequence neural networks. In: ICLR (2016). https:\/\/doi.org\/10.1103\/PhysRevLett.116.082003. arXiv:1511.05493","DOI":"10.1103\/PhysRevLett.116.082003"},{"key":"5484_CR18","unstructured":"Liu, Q., Allamanis, M., Brockschmidt, M., Gaunt, A.L.: Constrained graph variational autoencoders for molecule design. In: NeurIPS (2018). arXiv:1805.09076"},{"issue":"3","key":"5484_CR19","doi-asserted-by":"publisher","first-page":"498","DOI":"10.1109\/TNN.2008.2010350","volume":"20","author":"A Micheli","year":"2009","unstructured":"Micheli A (2009) Neural network for graphs: a contextual constructive approach. IEEE Trans. Neural Netw. 20(3):498\u2013511. https:\/\/doi.org\/10.1109\/TNN.2008.2010350","journal-title":"IEEE Trans. Neural Netw."},{"key":"5484_CR20","doi-asserted-by":"publisher","unstructured":"Morris C, Ritzert M, Fey M, Hamilton WL, Lenssen JE, Rattan G, Grohe M (2019) Weisfeiler and Leman Go Neural: higher-order graph neural networks. In: Proceedings of the AAAI conference on artificial intelligence, vol. 33, pp 4602\u20134609. https:\/\/doi.org\/10.1609\/aaai.v33i01.33014602. arXiv:1810.02244","DOI":"10.1609\/aaai.v33i01.33014602"},{"key":"5484_CR21","doi-asserted-by":"crossref","unstructured":"Navarin N, Tran DV, Sperduti A (2019) Universal readout for graph convolutional neural networks. In: International joint conference on neural networks. Budapest, Hungary","DOI":"10.1109\/IJCNN.2019.8852103"},{"key":"5484_CR22","unstructured":"Navarin N, Tran DV, Sperduti A (2020) Learning kernel-based embeddings in graph neural networks. In: European conference on artificial intelligence"},{"key":"5484_CR23","unstructured":"Niepert M, Ahmed M, Kutzkov K (2016) Learning convolutional neural networks for graphs. In: ICML, pp 2014\u20132023"},{"issue":"1","key":"5484_CR24","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","volume":"20","author":"F Scarselli","year":"2009","unstructured":"Scarselli F, Gori M, Ah Chung Tsoi AC, Hagenbuchner M, Monfardini G (2009) The graph neural network model. IEEE Trans Neural Netw 20(1):61\u201380. https:\/\/doi.org\/10.1109\/TNN.2008.2005605","journal-title":"IEEE Trans Neural Netw"},{"issue":"77","key":"5484_CR25","first-page":"2539","volume":"12","author":"N Shervashidze","year":"2011","unstructured":"Shervashidze N, Schweitzer P, Van Leeuwen EJ, Mehlhorn K, Borgwardt KM (2011) Weisfeiler\u2013Lehman graph kernels. J Mach Learn Res 12(77):2539\u20132561","journal-title":"J Mach Learn Res"},{"key":"5484_CR26","doi-asserted-by":"crossref","unstructured":"Simonovsky M, Komodakis N (2017) Dynamic edge-conditioned filters in convolutional neural networks on graphs. arxiv:1704.02901. Comment: Accepted to CVPR 2017; extended version","DOI":"10.1109\/CVPR.2017.11"},{"issue":"3","key":"5484_CR27","doi-asserted-by":"publisher","first-page":"714","DOI":"10.1109\/72.572108","volume":"8","author":"A Sperduti","year":"1997","unstructured":"Sperduti A, Starita A (1997) Supervised neural networks for the classification of structures. IEEE Trans Neural Netw 8(3):714\u2013735. https:\/\/doi.org\/10.1109\/72.572108","journal-title":"IEEE Trans Neural Netw"},{"key":"5484_CR28","doi-asserted-by":"publisher","unstructured":"Tran DV, Navarin N, Sperduti A (2018) On filter size in graph convolutional networks. In: IEEE SSCI. IEEE, Bengaluru, pp 1534\u20131541. https:\/\/doi.org\/10.1109\/SSCI.2018.8628758. https:\/\/ieeexplore.ieee.org\/document\/8628758\/","DOI":"10.1109\/SSCI.2018.8628758"},{"issue":"3","key":"5484_CR29","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1007\/s10115-007-0103-5","volume":"14","author":"N Wale","year":"2008","unstructured":"Wale N, Watson IA, Karypis G (2008) Comparison of descriptor spaces for chemical compound retrieval and classification. Knowl Inf Syst 14(3):347\u2013375","journal-title":"Knowl Inf Syst"},{"key":"5484_CR30","unstructured":"Xu K, Hu W, Leskovec J, Jegelka S (2019) How powerful are graph neural networks? In: International conference on learning representations"},{"key":"5484_CR31","unstructured":"Ying Z, You J, Morris C, Ren X, Hamilton W, Leskovec J (2018) Hierarchical graph representation learning with differentiable pooling. In: Advances in neural information processing systems, pp 4800\u20134810"},{"key":"5484_CR32","unstructured":"Zaheer M, Kottur S, Ravanbhakhsh S, P\u00f3czos B, Salakhutdinov R, Smola AJ (2017) Deep sets. In: Advances in neural information processing systems, pp 3391\u20133401 (2017)"},{"key":"5484_CR33","doi-asserted-by":"crossref","unstructured":"Zhang M, Cui Z, Neumann M, Chen Y (2018) An end-to-end deep learning architecture for graph classification. In: AAAI conference on artificial intelligence","DOI":"10.1609\/aaai.v32i1.11782"},{"key":"5484_CR34","doi-asserted-by":"crossref","unstructured":"Zhang M, Cui Z, Neumann M, Chen Y (2018) An end-to-end deep learning architecture for graph classification. In: Thirty-second AAAI conference on artificial intelligence","DOI":"10.1609\/aaai.v32i1.11782"}],"container-title":["Neural Computing and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-020-05484-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00521-020-05484-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-020-05484-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,11,28]],"date-time":"2022-11-28T13:43:10Z","timestamp":1669642990000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00521-020-05484-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,11,17]]},"references-count":34,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2022,1]]}},"alternative-id":["5484"],"URL":"https:\/\/doi.org\/10.1007\/s00521-020-05484-4","relation":{},"ISSN":["0941-0643","1433-3058"],"issn-type":[{"value":"0941-0643","type":"print"},{"value":"1433-3058","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,11,17]]},"assertion":[{"value":"13 May 2020","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 October 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 November 2020","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Compliance with ethical standard"}},{"value":"The authors declare that they have no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}