{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T12:58:01Z","timestamp":1781355481409,"version":"3.54.1"},"reference-count":33,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2022,6,29]],"date-time":"2022-06-29T00:00:00Z","timestamp":1656460800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,6,29]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Graph embedding is a transformation of nodes of a graph into a set of vectors. A good embedding should capture the graph topology, node-to-node relationship and other relevant information about the graph, its subgraphs and nodes. If these objectives are achieved, an embedding is a meaningful, understandable, compressed representations of a network that can be used for other machine learning tools such as node classification, community detection or link prediction. In this article, we do a series of extensive experiments with selected graph embedding algorithms, both on real-world networks as well as artificially generated ones. Based on those experiments, we formulate the following general conclusions. First, we confirm the main problem of node embeddings that is rather well-known to practitioners but less documented in the literature. There exist many algorithms available to choose from which use different techniques and have various parameters that may be tuned, the dimension being one of them. One needs to ensure that embeddings describe the properties of the underlying graphs well but, as our experiments confirm, it highly depends on properties of the network at hand and the given application in mind. As a result, selecting the best embedding is a challenging task and very often requires domain experts. Since investigating embeddings in a supervised manner is computationally expensive, there is a need for an unsupervised tool that is able to select a handful of promising embeddings for future (supervised) investigation. A general framework, introduced recently in the literature and easily available on GitHub repository, provides one of the very first tools for an unsupervised graph embedding comparison by assigning the \u2018divergence score\u2019 to embeddings with a goal of distinguishing good from bad ones. We show that the divergence score strongly correlates with the quality of embeddings by investigating three main applications of node embeddings: node classification, community detection and link prediction.<\/jats:p>","DOI":"10.1093\/comnet\/cnac030","type":"journal-article","created":{"date-parts":[[2022,8,4]],"date-time":"2022-08-04T07:22:36Z","timestamp":1659597756000},"source":"Crossref","is-referenced-by-count":15,"title":["Evaluating node embeddings of complex networks"],"prefix":"10.1093","volume":"10","author":[{"given":"Arash","family":"Dehghan-Kooshkghazi","sequence":"first","affiliation":[{"name":"Department of Mathematics, Toronto Metropolitan University , Toronto, ON M5B 2K3, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bogumi\u0142","family":"Kami\u0144ski","sequence":"additional","affiliation":[{"name":"Decision Analysis and Support Unit, SGH Warsaw School of Economics , Warsaw 02-554, Poland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"\u0141ukasz","family":"Krai\u0144ski","sequence":"additional","affiliation":[{"name":"Decision Analysis and Support Unit, SGH Warsaw School of Economics , Warsaw 02-554, Poland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pawe\u0142","family":"Pra\u0142at","sequence":"additional","affiliation":[{"name":"Department of Mathematics, Toronto Metropolitan University , Toronto, ON M5B 2K3, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fran\u00e7ois","family":"Th\u00e9berge","sequence":"additional","affiliation":[{"name":"Tutte Institute for Mathematics and Computing , Ottawa, ON K1J 0B9, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,8,4]]},"reference":[{"key":"2022080407211617500_B1","doi-asserted-by":"crossref","DOI":"10.1093\/acprof:oso\/9780199206650.001.0001","volume-title":"Networks: An Introduction","author":"Newman,","year":"2010"},{"key":"2022080407211617500_B2","volume-title":"Network Science","author":"Barabasi,","year":"2016"},{"key":"2022080407211617500_B3","volume-title":"Mining Complex Networks.","author":"Kami\u0144ski,","year":"2022"},{"key":"2022080407211617500_B4","doi-asserted-by":"crossref","first-page":"56001","DOI":"10.1209\/0295-5075\/111\/56001","article-title":"Interdisciplinary and physics challenges of network theory","volume":"111","author":"Bianconi,","year":"2015","journal-title":"Eur. Phys. Lett."},{"key":"2022080407211617500_B5","doi-asserted-by":"crossref","first-page":"1090","DOI":"10.1198\/016214502388618906","article-title":"Latent space approaches to social network analysis","volume":"97","author":"Hoff,","year":"2002","journal-title":"J. Am. Stat. Assoc."},{"key":"2022080407211617500_B6","first-page":"1","article-title":"Clustering means geometry in networks","volume":"208302","author":"Krioukov,","year":"2016","journal-title":"Phys Rev Lett."},{"key":"2022080407211617500_B7","doi-asserted-by":"crossref","first-page":"9421","DOI":"10.1038\/srep09421","article-title":"Emergence of soft communities from geometric preferential attachment","volume":"5","author":"Zuev,","year":"2015","journal-title":"Sci. Rep."},{"key":"2022080407211617500_B8","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1140\/epjb\/e2006-00046-8","article-title":"The spatial structure of networks","volume":"49","author":"Gastner,","year":"2006","journal-title":"Eur. Phys. J. B"},{"key":"2022080407211617500_B9","doi-asserted-by":"crossref","first-page":"7663","DOI":"10.1073\/pnas.1018962108","article-title":"Uncovering space-independent communities in spatial networks","volume":"108","author":"Expert,","year":"2011","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"2022080407211617500_B10","first-page":"201","article-title":"Spatial models for virtual networks","author":"Janssen,","year":"2010","journal-title":"CiE 2010, LNCS 6158."},{"key":"2022080407211617500_B11","doi-asserted-by":"crossref","first-page":"e357","DOI":"10.7717\/peerj-cs.357","article-title":"Survey on graph embeddings and their applications to machine learning problems on graphs","volume":"7","author":"Makarov,","year":"2021","journal-title":"PeerJ Comput. Sci."},{"key":"2022080407211617500_B12","first-page":"52","article-title":"Representation learning on graphs: methods and applications","volume":"40","author":"Hamilton,","year":"2017","journal-title":"IEEE Data Eng. Bull."},{"key":"2022080407211617500_B13","article-title":"Inductive representation learning on large graphs","author":"Hamilton,","year":"2017","journal-title":"arXiv:1706.02216 [cs.SI]"},{"key":"2022080407211617500_B14","doi-asserted-by":"crossref","first-page":"1465","DOI":"10.1007\/s10994-020-05890-8","article-title":"Propositionalization and embeddings: two sides of the same coin","volume":"109","author":"Lavra\u010d,","year":"2020","journal-title":"Mach Learn."},{"key":"2022080407211617500_B15","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","article-title":"Silhouettes: a graphical aid to the interpretation and validation of cluster analysis","volume":"20","author":"Rousseeuw,","year":"1987","journal-title":"Comput. Appl. Math."},{"key":"2022080407211617500_B16","article-title":"An unsupervised framework for comparing graph embeddings","volume":"8","author":"Kami\u0144ski,","year":"2020","journal-title":"J. Complex Netw."},{"key":"2022080407211617500_B17","first-page":"52","article-title":"A scalable unsupervised framework for comparing graph embeddings","volume-title":"Proceedings of the 17th Workshop on Algorithms and Models for the Web Graph (WAW 2020), Lecture Notes in Computer Science 12091","author":"Kami\u0144ski,","year":"2020"},{"key":"2022080407211617500_B18","doi-asserted-by":"crossref","first-page":"78","DOI":"10.1016\/j.knosys.2018.03.022","article-title":"Graph embedding techniques, applications, and performance: a survey","volume":"151","author":"Goyal,","year":"2018","journal-title":"Knowl. Based Syst."},{"key":"2022080407211617500_B19","article-title":"DeepWalk: online learning of social representations","author":"Perozzi,","year":"2014","journal-title":"KDD"},{"key":"2022080407211617500_B20","doi-asserted-by":"crossref","first-page":"855","DOI":"10.1145\/2939672.2939754","article-title":"node2vec: scalable feature learning for networks","author":"Grover,","year":"2016","journal-title":"KDD."},{"key":"2022080407211617500_B21","article-title":"Distributed representations of words and phrases and their compositionality","author":"Mikolov,","year":"2013","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"2022080407211617500_B22","first-page":"1225","article-title":"Structural deep network embedding","author":"Wang,","year":"2016","journal-title":"Proceedings of ACM SIGKDD."},{"key":"2022080407211617500_B23","doi-asserted-by":"crossref","DOI":"10.1145\/2939672.2939751","article-title":"Asymmetric transitivity preserving graph embedding","author":"Ou,","year":"2016","journal-title":"KDD."},{"key":"2022080407211617500_B24","doi-asserted-by":"crossref","first-page":"1067","DOI":"10.1145\/2736277.2741093","article-title":"Line: large-scale information network embedding","author":"Tang,","year":"2015","journal-title":"Proceedings 24th International Conference on World Wide Web."},{"key":"2022080407211617500_B25","first-page":"539","article-title":"VERSE: versatile graph embeddings from similarity measures","volume-title":"Proceedings of the 2018 World Wide Web Conference (WWW\u201918). International World Wide Web.","author":"Tsitsulin,","year":"2018"},{"key":"2022080407211617500_B26","article-title":"Ensemble clustering for graphs. Complex networks and their applications VII","volume-title":"COMPLEX NETWORKS 2018. Studies in Computational Intelligence","author":"Poulin,","year":"2018"},{"key":"2022080407211617500_B27","doi-asserted-by":"crossref","first-page":"7821","DOI":"10.1073\/pnas.122653799","article-title":"Community structure in social and biological networks","volume":"99","author":"Girvan,","year":"2002","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"2022080407211617500_B28","doi-asserted-by":"crossref","first-page":"5982","DOI":"10.1038\/s41598-018-23932-z","article-title":"Community detection in complex networks via clique conductance","volume":"8","author":"Lu,","year":"2018","journal-title":"Nat. Sci. Rep."},{"key":"2022080407211617500_B29","article-title":"UMAP: uniform manifold approximation and projection for dimension reduction","author":"McInnes,","year":"2018","journal-title":"pre-print arXiv:1802.03426"},{"key":"2022080407211617500_B30","doi-asserted-by":"crossref","DOI":"10.1103\/PhysRevE.78.046110","article-title":"Benchmark graphs for testing community detection algorithms","volume":"78","author":"Lancichinetti,","year":"2008","journal-title":"Phys. Rev. E"},{"key":"2022080407211617500_B31","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1017\/nws.2020.45","article-title":"Artificial benchmark for community detection (ABCD) \u2014 fast random graph model with community structure","volume":"9","author":"Kami\u0144ski,","year":"2021","journal-title":"Netw. Sci."},{"key":"2022080407211617500_B32","article-title":"Properties and performance of the ABCD random graph model with community structure","author":"Kami\u0144ski,","year":"2022","journal-title":"pre-print arXiv:2203.14899"},{"key":"2022080407211617500_B33","article-title":"Modularity of the ABCD random graph model with community structure","author":"Kami\u0144ski,","year":"2022","journal-title":"pre-print arXiv:2203.01480"}],"container-title":["Journal of Complex Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/comnet\/article-pdf\/10\/4\/cnac030\/45241730\/cnac030.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/comnet\/article-pdf\/10\/4\/cnac030\/45241730\/cnac030.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,4]],"date-time":"2022-08-04T07:23:09Z","timestamp":1659597789000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/comnet\/article\/doi\/10.1093\/comnet\/cnac030\/6655565"}},"subtitle":[],"editor":[{"given":"Ali","family":"Pinar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2022,6,29]]},"references-count":33,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2022,6,29]]}},"URL":"https:\/\/doi.org\/10.1093\/comnet\/cnac030","relation":{},"ISSN":["2051-1329"],"issn-type":[{"value":"2051-1329","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,8,1]]},"published":{"date-parts":[[2022,6,29]]},"article-number":"cnac030"}}