{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,9,11]],"date-time":"2024-09-11T05:54:18Z","timestamp":1726034058878},"reference-count":60,"publisher":"Association for Computing Machinery (ACM)","issue":"12","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2019,8]]},"abstract":"\n An increasing number of machine learning tasks require dealing with large graph datasets, which capture rich and complex relationship among potentially billions of elements. Graph Neural Network (GNN) becomes an effective way to address the graph learning problem by converting the graph data into a low dimensional space while keeping both the structural and property information to the maximum extent and constructing a neural network for training and referencing. However, it is challenging to provide an efficient graph storage and computation capabilities to facilitate GNN training and enable development of new GNN algorithms. In this paper, we present a comprehensive graph neural network system, namely\n AliGraph<\/jats:italic>\n , which consists of distributed graph storage, optimized sampling operators and runtime to efficiently support not only existing popular GNNs but also a series of in-house developed ones for different scenarios. The system is currently deployed at Alibaba to support a variety of business scenarios, including product recommendation and personalized search at Alibaba's E-Commerce platform. By conducting extensive experiments on a real-world dataset with 492.90 million vertices, 6.82 billion edges and rich attributes,\n AliGraph<\/jats:italic>\n performs an order of magnitude faster in terms of graph building (5 minutes vs hours reported from the state-of-the-art PowerGraph platform). At training,\n AliGraph<\/jats:italic>\n runs 40%-50% faster with the novel caching strategy and demonstrates around 12 times speed up with the improved runtime. In addition, our in-house developed GNN models all showcase their statistically significant superiorities in terms of both effectiveness and efficiency (e.g., 4.12%--17.19% lift by F1 scores).\n <\/jats:p>","DOI":"10.14778\/3352063.3352127","type":"journal-article","created":{"date-parts":[[2019,9,18]],"date-time":"2019-09-18T18:36:11Z","timestamp":1568831771000},"page":"2094-2105","source":"Crossref","is-referenced-by-count":149,"title":["AliGraph"],"prefix":"10.14778","volume":"12","author":[{"given":"Rong","family":"Zhu","sequence":"first","affiliation":[{"name":"Alibaba Group"}]},{"given":"Kun","family":"Zhao","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]},{"given":"Hongxia","family":"Yang","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]},{"given":"Wei","family":"Lin","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]},{"given":"Chang","family":"Zhou","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]},{"given":"Baole","family":"Ai","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]},{"given":"Yong","family":"Li","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]},{"given":"Jingren","family":"Zhou","sequence":"additional","affiliation":[{"name":"Alibaba Group"}]}],"member":"320","published-online":{"date-parts":[[2019,8]]},"reference":[{"key":"e_1_2_1_1_1","first-page":"4502","volume-title":"NIPS","author":"Battaglia P.","year":"2016"},{"issue":"3","key":"e_1_2_1_2_1","first-page":"115","article-title":"Node classification in social networks","volume":"16","author":"Bhagat S.","year":"2011","journal-title":"Computer Science"},{"key":"e_1_2_1_3_1","doi-asserted-by":"crossref","unstructured":"E. G. Boman K. D. Devine and S. Rajamanickam. Scalable matrix computations on large scale-free graphs using 2d graph partitioning. 2013. E. G. Boman K. D. Devine and S. Rajamanickam. Scalable matrix computations on large scale-free graphs using 2d graph partitioning. 2013.","DOI":"10.1145\/2503210.2503293"},{"key":"e_1_2_1_4_1","first-page":"568","article-title":"Experiments on graph clustering algorithms","volume":"2832","author":"Brandes U.","year":"2003","journal-title":"LNCS"},{"issue":"9","key":"e_1_2_1_5_1","first-page":"1616","article-title":"A comprehensive survey of graph embedding: Problems, techniques and applications","volume":"30","author":"Cai H.","year":"2017","journal-title":"TKDE"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/2783258.2783296"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330964"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330967"},{"key":"e_1_2_1_9_1","volume-title":"IJCAI","author":"H.","year":"2019"},{"key":"e_1_2_1_10_1","volume-title":"IJCAI","author":"Tang L.","year":"2019"},{"key":"e_1_2_1_11_1","unstructured":"J. Chen T. Ma and C. Xiao. Fastgcn: fast learning with graph convolutional networks via importance sampling. arXiv.1801.10247 2018. J. Chen T. Ma and C. Xiao. Fastgcn: fast learning with graph convolutional networks via importance sampling. arXiv.1801.10247 2018."},{"key":"e_1_2_1_12_1","volume-title":"Acm-siam Symposium on Discrete Algorithms","author":"Chrobak M.","year":"1998"},{"key":"e_1_2_1_13_1","volume-title":"TKDE","author":"Cui P.","year":"2018"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/3097983.3098036"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.5555\/3304889.3304949"},{"key":"e_1_2_1_16_1","first-page":"5119","volume-title":"Advances in Neural Information Processing Systems","author":"Duran A. G.","year":"2017"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3035942"},{"key":"e_1_2_1_18_1","first-page":"6530","volume-title":"NIPS","author":"Fout A.","year":"2017"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.5555\/3304222.3304235"},{"key":"e_1_2_1_20_1","volume-title":"OSDI","author":"Gonzalez J. E.","year":"2012"},{"key":"e_1_2_1_21_1","volume-title":"Knowledge-Based Systems","author":"Goyal P.","year":"2018"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939754"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.5555\/3172077.3172138"},{"key":"e_1_2_1_24_1","unstructured":"W. L. Hamilton R. Ying and J. Leskovec. Representation learning on graphs: Methods and applications. 2017. W. L. Hamilton R. Ying and J. Leskovec. Representation learning on graphs: Methods and applications. 2017."},{"key":"e_1_2_1_25_1","first-page":"1025","volume-title":"NIPS","author":"Hamilton W. L.","year":"2017"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/2872427.2883037"},{"key":"e_1_2_1_27_1","first-page":"633","volume-title":"SDM","author":"Huang X.","year":"2017"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3018661.3018667"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICNN.1988.23877"},{"key":"e_1_2_1_31_1","unstructured":"D. P. Kingma and M. Welling. Auto-encoding variational bayes. arXiv.1312.6114 2013. D. P. Kingma and M. Welling. Auto-encoding variational bayes. arXiv.1312.6114 2013."},{"key":"e_1_2_1_32_1","volume-title":"ICLR","author":"Kipf T. N.","year":"2017"},{"key":"e_1_2_1_33_1","first-page":"521","volume-title":"Nature","author":"LeCun Y.","year":"2015"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3132847.3132919"},{"key":"e_1_2_1_35_1","volume-title":"IJCAI","author":"H.","year":"2019"},{"key":"e_1_2_1_36_1","doi-asserted-by":"crossref","unstructured":"D. Liang R. G. Krishnan M. D. Hoffman and T. Jebara. Variational autoencoders for collaborative filtering. 2018. D. Liang R. G. Krishnan M. D. Hoffman and T. Jebara. Variational autoencoders for collaborative filtering. 2018.","DOI":"10.1145\/3178876.3186150"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2018.2819980"},{"key":"e_1_2_1_38_1","doi-asserted-by":"crossref","unstructured":"D. Liben-Nowell and J. Kleinberg. The link prediction problem for social networks. 2003. D. Liben-Nowell and J. Kleinberg. The link prediction problem for social networks. 2003.","DOI":"10.1145\/956863.956972"},{"key":"e_1_2_1_39_1","unstructured":"Z. Lin M. Feng C. N. d. Santos M. Yu B. Xiang B. Zhou and Y. Bengio. A structured self-attentive sentence embedding. arXiv:1703.03130 2017. Z. Lin M. Feng C. N. d. Santos M. Yu B. Xiang B. Zhou and Y. Bengio. A structured self-attentive sentence embedding. arXiv:1703.03130 2017."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDMW.2017.23"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330967"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/2766462.2767755"},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1145\/2623330.2623732"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/3159652.3159706"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3132847.3133021"},{"key":"e_1_2_1_46_1","unstructured":"L. F. R. Ribeiro P. H. P. Saverese and D. R. Figueiredo. struc2vec : Learning node representations from structural identity. 2017. L. F. R. Ribeiro P. H. P. Saverese and D. R. Figueiredo. struc2vec : Learning node representations from structural identity. 2017."},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2018.2833443"},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.1145\/2339530.2339722"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/2783258.2783307"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1145\/2736277.2741093"},{"key":"e_1_2_1_51_1","volume-title":"Physics","author":"Tanimoto S.","year":"2009"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1145\/1390156.1390294"},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939753"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1109\/APWeb.2010.60"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-1223"},{"key":"e_1_2_1_56_1","volume-title":"IJCAI","author":"Yang C.","year":"2015"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.5555\/3304889.3305089"},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.5555\/3304889.3305099"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDM.2018.00198"},{"key":"e_1_2_1_60_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11257"},{"key":"e_1_2_1_61_1","volume-title":"IJCAI","author":"Tang L.","year":"2019"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3352063.3352127","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,28]],"date-time":"2022-12-28T10:42:29Z","timestamp":1672224149000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3352063.3352127"}},"subtitle":["a comprehensive graph neural network platform"],"short-title":[],"issued":{"date-parts":[[2019,8]]},"references-count":60,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2019,8]]}},"alternative-id":["10.14778\/3352063.3352127"],"URL":"https:\/\/doi.org\/10.14778\/3352063.3352127","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2019,8]]}}}