Node Classification On Pubmed
المقاييس
Accuracy
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | Accuracy |
|---|---|
| deep-autoencoder-like-nonnegative-matrix | 63.93% |
| cleora-a-simple-strong-and-scalable-graph | 80.2 |
| measuring-and-relieving-the-over-smoothing | 77.4 ± 0.2 |
| understanding-over-squashing-and-bottlenecks-1 | 79.10±0.11 |
| multi-mask-aggregators-for-graph-neural | 86.00% |
| inferring-from-references-with-differences | 88.90 |
| clarify-confused-nodes-through-separated | 91.55 ± 0.38 |
| predict-then-propagate-graph-neural-networks | 79.73 ± 0.31 |
| gresnet-graph-residuals-for-reviving-deep | 81.7% |
| gresnet-graph-residuals-for-reviving-deep | 82.2% |
| distilling-self-knowledge-from-contrastive | 88.86% |
| unifying-graph-convolutional-neural-networks-1 | 87.8 ± 0.6 |
| auto-gnn-neural-architecture-search-of-graph | 79.7 ± 0.4% |
| fast-graph-representation-learning-with | 79.4 ± 2.2 |
| convolutional-neural-networks-on-graphs-with | 74.4% |
| revisiting-semi-supervised-learning-with | 77.2% |
| clarify-confused-nodes-through-separated | 91.64 ± 0.53 |
| graph-bert-only-attention-is-needed-for | 79.3% |
| n-gcn-multi-scale-graph-convolution-for-semi | 79.5% |
| is-heterophily-a-real-nightmare-for-graph | 90.74 ± 0.5 |
| learning-to-make-predictions-on-graphs-with | 79.40% |
| mixhop-higher-order-graph-convolution | 80.8% |
| the-split-matters-flat-minima-methods-for | 90.64 ± 0.46% |
| tackling-oversmoothing-of-gnns-with-1 | 81.92±0.13 |
| is-heterophily-a-real-nightmare-for-graph | 91.31 ± 0.6 |
| cluster-gcn-an-efficient-algorithm-for | - |
| graph-star-net-for-generalized-multi-task-1 | 77.2% |
| graph-representation-learning-beyond-node-and | - |
| optimization-of-graph-neural-networks-with | 89.36 ± 0.57 |
| splinecnn-fast-geometric-deep-learning-with | 88.88% |
| is-heterophily-a-real-nightmare-for-graph | 90.56 ± 0.39 |
| diffwire-inductive-graph-rewiring-via-the | 68.19 |
| deep-graph-infomax | 76.8 ± 0.6% |
| distilling-self-knowledge-from-contrastive | 88.79% |
| graphnas-graph-neural-architecture-search | 79.6 ± 0.4% |
| النموذج 36 | 81.9% |
| semi-supervised-classification-with-graph | 79.0 |
| graph-attention-networks | 79.0 ± 0.3% |
| structure-fusion-based-on-graph-convolutional | 80.0% |
| cn-motifs-perceptive-graph-neural-networks | 90.07± 0.43 |
| a-flexible-generative-framework-for-graph | 78.4% |
| adagcn-adaboosting-graph-convolutional | 79.76 ± 0.27 |
| nodenet-a-graph-regularised-neural-network | 90.21% |
| graph-infoclust-leveraging-cluster-level-node | 77.4 ± 1.9 |
| a-capsule-network-based-model-for-learning | 78.45% |
| graph-trees-with-attention | 78.0 |
| mitigating-degree-biases-in-message-passing | 86.86±0.12 |
| graph-less-neural-networks-teaching-old-mlps-1 | 75.42 ± 2.31 |
| diffwire-inductive-graph-rewiring-via-the | 86.07 |
| large-scale-learnable-graph-convolutional | 79.5% |
| beyond-homophily-structure-aware-path | - |
| dfnets-spectral-cnns-for-graphs-with-feedback | 85.2 ± 0.3 |
| structure-fusion-based-on-graph-convolutional | 79.3% |
| graph-optimized-convolutional-networks | 79.7% |
| graphite-iterative-generative-modeling-of | 79.3 ± 0.03 |
| certifiable-robustness-and-robust-training | 86% |
| multi-task-graph-autoencoders | 80.40% |
| deep-graph-contrastive-representation | 86.7 ± 0.1 |
| graph-wavelet-neural-network-1 | 79.1% |
| gresnet-graph-residuals-for-reviving-deep | 83.0% |
| mixup-for-node-and-graph-classification | 87.9% |
| graph-u-nets | 79.6 ± 0.2% |
| distilling-self-knowledge-from-contrastive | 89.58% |
| get-rid-of-suspended-animation-problem-deep | 79.5% |
| gresnet-graph-residuals-for-reviving-deep | 81.2% |
| distilling-self-knowledge-from-contrastive | 89.53% |
| diffusion-improves-graph-learning-1 | 79.95% |
| 190807558 | 82.92 ± 0.13 |
| hyperbolic-graph-convolutional-neural | 80.3% |