{"url":"/sota/node-classification-on-pattern-100k","task":{"name":"Node Classification","url":"/task/node-classification","note":null},"dataset":{"name":"PATTERN 100k","url":null},"category":"Graphs","categories":["Graphs"],"category_note":null,"description":"**Node Classification** is a machine learning task in graph-based data analysis, where the goal is to assign labels to nodes in a graph based on the properties of nodes and the relationships between them.\r\n\r\n**Node Classification** models aim to predict non-existing node properties (known as the target property) based on other node properties. Typical models used for node classification consists of a large family of graph neural networks. Model performance can be measured using benchmark datasets like [Cora](/dataset/cora), [Citeseer](/dataset/citeseer), and [Pubmed](/dataset/pubmed), among others, typically using Accuracy and F1.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Fast Graph Representation Learning With PyTorch Geometric](https://arxiv.org/pdf/1903.02428v3.pdf) )</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Accuracy (%)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Accuracy (%)":"higher"}},"counts":{"rows":9,"rows_with_code":9,"rows_with_paper_page":9,"rows_dated":9,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"EGT","metrics":{"Accuracy (%)":"86.816"},"uses_additional_data":false,"paper_date":"2021-08-07","paper":"/paper/edge-augmented-graph-transformers-global-self","paper_url":"https://arxiv.org/abs/2108.03348v3","paper_title":"Global Self-Attention as a Replacement for Graph Convolution","code":"https://github.com/shamim-hussain/egt_pytorch","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"DGN","metrics":{"Accuracy (%)":"86.680"},"uses_additional_data":false,"paper_date":"2020-10-06","paper":"/paper/directional-graph-networks-1","paper_url":"https://arxiv.org/abs/2010.02863v4","paper_title":"Directional Graph Networks","code":"https://github.com/Saro00/DGN","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":8,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"FactorGCN","metrics":{"Accuracy (%)":"86.57 ± 0.02"},"uses_additional_data":false,"paper_date":"2020-10-12","paper":"/paper/factorizable-graph-convolutional-networks","paper_url":"https://arxiv.org/abs/2010.05421v1","paper_title":"Factorizable Graph Convolutional Networks","code":"https://github.com/ihollywhy/FactorGCN.PyTorch","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"PNA","metrics":{"Accuracy (%)":"86.567"},"uses_additional_data":false,"paper_date":"2020-04-12","paper":"/paper/principal-neighbourhood-aggregation-for-graph","paper_url":"https://arxiv.org/abs/2004.05718v5","paper_title":"Principal Neighbourhood Aggregation for Graph Nets","code":"https://github.com/rusty1s/pytorch_geometric","n_code_links":8,"syntology":{"n_ran":33,"n_unverified":22,"n_samples":55,"n_pointer_only_licence":48}},{"rank_in_archive_order":5,"model":"GIN","metrics":{"Accuracy (%)":"85.590"},"uses_additional_data":false,"paper_date":"2018-10-01","paper":"/paper/how-powerful-are-graph-neural-networks","paper_url":"http://arxiv.org/abs/1810.00826v3","paper_title":"How Powerful are Graph Neural Networks?","code":"https://github.com/dmlc/dgl/tree/master/examples/mxnet/gin","n_code_links":19,"syntology":{"n_ran":3,"n_unverified":7,"n_samples":10,"n_pointer_only_licence":5}},{"rank_in_archive_order":6,"model":"MoNet","metrics":{"Accuracy (%)":"85.482"},"uses_additional_data":false,"paper_date":"2016-11-25","paper":"/paper/geometric-deep-learning-on-graphs-and","paper_url":"http://arxiv.org/abs/1611.08402v3","paper_title":"Geometric deep learning on graphs and manifolds using mixture model CNNs","code":"https://github.com/dmlc/dgl/tree/master/examples/mxnet/monet","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"GatedGCN","metrics":{"Accuracy (%)":"84.480"},"uses_additional_data":false,"paper_date":"2017-11-20","paper":"/paper/residual-gated-graph-convnets","paper_url":"http://arxiv.org/abs/1711.07553v2","paper_title":"Residual Gated Graph ConvNets","code":"https://github.com/xbresson/spatial_graph_convnets","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"GAT","metrics":{"Accuracy (%)":"75.824"},"uses_additional_data":false,"paper_date":"2017-10-30","paper":"/paper/graph-attention-networks","paper_url":"http://arxiv.org/abs/1710.10903v3","paper_title":"Graph Attention Networks","code":"https://github.com/labmlai/annotated_deep_learning_paper_implementations","n_code_links":93,"syntology":{"n_ran":50,"n_unverified":56,"n_samples":106,"n_pointer_only_licence":43}},{"rank_in_archive_order":9,"model":"GraphSage","metrics":{"Accuracy (%)":"50.516"},"uses_additional_data":false,"paper_date":"2017-06-07","paper":"/paper/inductive-representation-learning-on-large","paper_url":"http://arxiv.org/abs/1706.02216v4","paper_title":"Inductive Representation Learning on Large Graphs","code":"https://github.com/pyg-team/pytorch_geometric/blob/master/torch_geometric/nn/models/basic_gnn.py","n_code_links":20,"syntology":{"n_ran":3,"n_unverified":2,"n_samples":5,"n_pointer_only_licence":5}}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":8,"rows_with_any_sample_ran":6,"distinct_papers_with_graph_line":8,"distinct_papers_with_any_sample_ran":6,"samples_over_distinct_papers":{"n_ran":97,"n_unverified":104,"n_samples":201,"n_pointer_only_licence":101,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":97,"n_unverified":104,"n_samples":201,"n_pointer_only_licence":101,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}