{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T13:22:54Z","timestamp":1777555374125,"version":"3.51.4"},"reference-count":36,"publisher":"SAGE Publications","issue":"1-2","license":[{"start":{"date-parts":[[2017,10,17]],"date-time":"2017-10-17T00:00:00Z","timestamp":1508198400000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Data Science"],"published-print":{"date-parts":[[2017,12,8]]},"abstract":"<jats:p>In modern machine learning, raw data is the preferred input for our models. Where a decade ago data scientists were still engineering features, manually picking out the details we thought salient, they now prefer the data in their raw form. As long as we can assume that all relevant and irrelevant information is present in the input data, we can design deep models that build up intermediate representations to sift out relevant features. However, these models are often domain specific and tailored to the task at hand, and therefore unsuited for learning on heterogeneous knowledge: information of different types and from different domains. If we can develop methods that operate on this form of knowledge, we can dispense with a great deal more ad-hoc feature engineering and train deep models end-to-end in many more domains. To accomplish this, we first need a data model capable of expressing heterogeneous knowledge naturally in various domains, in as usable a form as possible, and satisfying as many use cases as possible. In this position paper, we argue that the knowledge graph is a suitable candidate for this data model. We further describe current research and discuss some of the promises and challenges of this approach.<\/jats:p>","DOI":"10.3233\/ds-170007","type":"journal-article","created":{"date-parts":[[2017,7,14]],"date-time":"2017-07-14T13:39:43Z","timestamp":1500039583000},"page":"39-57","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":55,"title":["The knowledge graph as the default data model for learning on heterogeneous knowledge"],"prefix":"10.1177","volume":"1","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2415-8438","authenticated-orcid":false,"given":"Xander","family":"Wilcke","sequence":"first","affiliation":[{"name":"Faculty of Sciences, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands."},{"name":"Faculty of Spatial Economics, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0189-5817","authenticated-orcid":false,"given":"Peter","family":"Bloem","sequence":"additional","affiliation":[{"name":"Faculty of Sciences, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9079-039X","authenticated-orcid":false,"given":"Victor","family":"de Boer","sequence":"additional","affiliation":[{"name":"Faculty of Sciences, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands."}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","published-online":{"date-parts":[[2017,10,17]]},"reference":[{"key":"ref001","unstructured":"A.\u00a0Abele, J.P.\u00a0McCrae, P.\u00a0Buitelaar, A.\u00a0Jentzsch and R.\u00a0Cyganiak, Linking open data cloud diagram. http:\/\/lod-cloud.net, Accessed: 2017-03-01."},{"key":"ref002","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-76298-0_3"},{"key":"ref003","unstructured":"A.\u00a0Bordes, N.\u00a0Usunier, A.\u00a0Garcia-Duran, J.\u00a0Weston and O.\u00a0Yakhnenko, Translating embeddings for modeling multi-relational data, in: Advances in Neural Information Processing Systems, 2013, pp.\u00a02787\u20132795, http:\/\/dl.acm.org\/citation.cfm?id=2999923."},{"key":"ref004","unstructured":"B.\u00a0Boser LeCun, J.S.\u00a0Denker, D.\u00a0Henderson, R.E.\u00a0Howard, W.\u00a0Hubbard and L.D.\u00a0Jackel, Handwritten digit recognition with a back-propagation network, in: Advances in Neural Information Processing Systems, Citeseer, 1990, http:\/\/dl.acm.org\/citation.cfm?id=2969879."},{"key":"ref005","unstructured":"L.\u00a0Bottou, Two big challenges in machine learning, http:\/\/icml.cc\/2015\/invited\/LeonBottouICML2015.pdf, Accessed: 2017-03-01."},{"key":"ref006","unstructured":"J.\u00a0Bruna, W.\u00a0Zaremba, A.\u00a0Szlam and Y.\u00a0LeCun, Spectral networks and locally connected networks on graphs,\n                      CoRR\n                      , arXiv preprint arXiv:1312.6203, 2013."},{"key":"ref007","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-31782-8_2"},{"key":"ref008","unstructured":"R.\u00a0Davis, H.\u00a0Shrobe and P.\u00a0Szolovits, What is a knowledge representation? AI magazine14(1) (1993), 17, http:\/\/citeseerx.ist.psu.edu\/viewdoc\/summary?doi=10.1.1.216.9376."},{"key":"ref009","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-04930-9_14"},{"key":"ref010","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638947"},{"key":"ref011","unstructured":"W.L.\u00a0Hamilton, R.\u00a0Ying and J.\u00a0Leskovec, Inductive representation learning on large graphs, arXiv preprint arXiv:1706.02216, 2017."},{"key":"ref012","doi-asserted-by":"publisher","DOI":"10.1109\/CSC.2011.6138544"},{"key":"ref013","doi-asserted-by":"crossref","unstructured":"M.A.\u00a0Hern\u00e1ndez and S.J.\u00a0Stolfo, The merge\/purge problem for large databases, in: ACM Sigmod Record, Vol.\u00a024, ACM, 1995, pp.\u00a0127\u2013138, http:\/\/dl.acm.org\/citation.cfm?id=223807.","DOI":"10.1145\/568271.223807"},{"key":"ref014","unstructured":"D.D.\u00a0Johnson, Learning graphical state transitions, in: Proceedings of the International Conference on Learning Representations (ICLR), 2017, https:\/\/openreview.net\/forum?id=HJ0NvFzxl."},{"key":"ref015","unstructured":"T.N.\u00a0Kipf and M.\u00a0Welling, Semi-supervised classification with graph convolutional networks,\n                      CoRR\n                      , arXiv preprint arXiv:1609.02907, 2016."},{"key":"ref016","doi-asserted-by":"publisher","DOI":"10.1016\/j.websem.2015.08.002"},{"key":"ref017","doi-asserted-by":"publisher","DOI":"10.1109\/MC.2009.263"},{"key":"ref018","unstructured":"A.\u00a0Krizhevsky, I.\u00a0Sutskever and G.E.\u00a0Hinton, Imagenet classification with deep convolutional neural networks, in: Advances in Neural Information Processing Systems, 2012, pp.\u00a01097\u20131105, http:\/\/dl.acm.org\/citation.cfm?id=3065386."},{"key":"ref019","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6639343"},{"key":"ref020","doi-asserted-by":"crossref","unstructured":"D.G.\u00a0Lowe, Object recognition from local scale-invariant features, in: Proceedings of the Seventh IEEE International Conference on Computer Vision, 1999, Vol.\u00a02, IEEE, 1999, pp.\u00a01150\u20131157, http:\/\/dl.acm.org\/citation.cfm?id=851523.","DOI":"10.1109\/ICCV.1999.790410"},{"key":"ref021","unstructured":"T.\u00a0Mikolov, K.\u00a0Chen, G.\u00a0Corrado and J.\u00a0Dean, Efficient estimation of word representations in vector space,\n                      CoRR\n                      , arXiv preprint arXiv:1301.3781, 2013."},{"key":"ref022","unstructured":"P.\u00a0Mineiro, Software engineering vs machine learning concepts, http:\/\/www.machinedlearnings.com\/2017\/02\/software-engineering-vs-machine.html, Accessed: 2017-03-01."},{"key":"ref023","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/W15-1506"},{"key":"ref024","unstructured":"M.\u00a0Nickel, V.\u00a0Tresp and H.P.\u00a0Kriegel, A three-way model for collective learning on multi-relational data, in: Proceedings of the 28th International Conference on Machine Learning (ICML-11), 2011, pp.\u00a0809\u2013816, http:\/\/dl.acm.org\/citation.cfm?id=3104584."},{"key":"ref025","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-34129-3_9"},{"key":"ref026","doi-asserted-by":"publisher","DOI":"10.1145\/2623330.2623732"},{"key":"ref027","unstructured":"T.\u00a0Pham, T.\u00a0Tran, D.Q.\u00a0Phung and S.\u00a0Venkatesh, Column networks for collective classification, in: Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence, San Francisco, California, USA, February 4\u20139, 2017, S.P.\u00a0Singh and S.\u00a0Markovitch, eds, AAAI Press, 2017, pp.\u00a02485\u20132491, arXiv preprint arXiv:1609.04508."},{"key":"ref028","unstructured":"Y.\u00a0Raimond and S.\u00a0Abdallah, The Timeline Ontology. OWL-DL Ontology, 2006, http:\/\/purl.org\/NET\/c4dm\/timeline.owl."},{"key":"ref029","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46523-4_30"},{"key":"ref030","doi-asserted-by":"crossref","unstructured":"M.\u00a0Schlichtkrull, T.N.\u00a0Kipf, P.\u00a0Bloem, R.\u00a0van den Berg, I.\u00a0Titov and M.\u00a0Welling, Modeling relational data with graph convolutional networks, arXiv preprint arXiv:1703.06103, 2017.","DOI":"10.1007\/978-3-319-93417-4_38"},{"key":"ref031","unstructured":"J.F.\u00a0Sequeda and O.\u00a0Corcho, Linked stream data: A position paper, in: Proceedings of the 2nd International Conference on Semantic Sensor Networks, Vol.\u00a0522, CEUR-WS.org, 2009, pp.\u00a0148\u2013157, http:\/\/dl.acm.org\/citation.cfm?id=2889944."},{"key":"ref032","first-page":"2539","volume":"12","author":"Shervashidze N.","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"ref033","doi-asserted-by":"crossref","unstructured":"B.\u00a0Shi and T.\u00a0Weninger, Proje: Embedding projection for knowledge graph completion, arXiv preprint arXiv:1611.05425, 2016.","DOI":"10.1609\/aaai.v31i1.10677"},{"key":"ref034","unstructured":"R.\u00a0Socher, D.\u00a0Chen, C.D.\u00a0Manning and A.\u00a0Ng, Reasoning with neural tensor networks for knowledge base completion, in: Advances in Neural Information Processing Systems, 2013, pp.\u00a0926\u2013934, http:\/\/citeseerx.ist.psu.edu\/viewdoc\/summary?doi=10.1.1.708.5787."},{"key":"ref035","doi-asserted-by":"crossref","unstructured":"R.\u00a0Socher, A.\u00a0Perelygin, J.Y.\u00a0Wu, J.\u00a0Chuang, C.D.\u00a0Manning, A.Y.\u00a0Ng, C.\u00a0Pottset al., Recursive deep models for semantic compositionality over a sentiment treebank, in: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), Vol.\u00a01631, Citeseer, 2013, p.\u00a01642, http:\/\/citeseerx.ist.psu.edu\/viewdoc\/summary?doi=10.1.1.593.7427.","DOI":"10.18653\/v1\/D13-1170"},{"key":"ref036","unstructured":"R.\u00a0van den Berg, T.N.\u00a0Kipf and M.\u00a0Welling, Graph convolutional matrix completion, arXiv preprint arXiv:1706.02263, 2017."}],"container-title":["Data Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/DS-170007","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.3233\/DS-170007","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/DS-170007","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T18:09:59Z","timestamp":1777399799000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.3233\/DS-170007"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,10,17]]},"references-count":36,"journal-issue":{"issue":"1-2","published-print":{"date-parts":[[2017,12,8]]}},"alternative-id":["10.3233\/DS-170007"],"URL":"https:\/\/doi.org\/10.3233\/ds-170007","relation":{},"ISSN":["2451-8484","2451-8492"],"issn-type":[{"value":"2451-8484","type":"print"},{"value":"2451-8492","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,10,17]]}}}