Graph Neural Networks (GNNs) have been widely applied to various fields for learning over graph-structured data. They have shown significant improvements over traditional heuristic methods in various tasks such as node classification and graph classification. However, since GNNs heavily rely on smoothed node features rather than graph structure, they often show poor performance than simple heuristic methods in link prediction where the structural information, e.g., overlapped neighborhoods, degrees, and shortest paths, is crucial. To address this limitation, we propose Neighborhood Overlap-aware Graph Neural Networks (Neo-GNNs) that learn useful structural features from an adjacency matrix and estimate overlapped neighborhoods for link prediction. Our Neo-GNNs generalize neighborhood overlap-based heuristic methods and handle overlapped multi-hop neighborhoods. Our extensive experiments on Open Graph Benchmark datasets (OGB) demonstrate that Neo-GNNs consistently achieve state-of-the-art performance in link prediction. Our code is publicly available at https://github.com/seongjunyun/Neo_GNNs.
翻译:神经网图(GNNs)被广泛应用于不同领域,用于通过图形结构数据学习,在诸如节点分类和图形分类等各种任务中,它们表明比传统的超光速方法有显著改进;然而,由于GNNS严重依赖平滑节点特征,而不是图形结构,因此在结构信息(例如相重叠的邻里、度和最短路径)至关重要的链接预测中,它们往往表现出比简单的超光速方法差的性能。为了应对这一限制,我们建议Neo-GNNNS(Neo-GNNS)从相邻矩阵中学习有用的结构特征,并估计相重叠的街区,以便进行链接预测。我们的Neo-GNNS普遍采用基于社区重叠的超光速方法,并处理重叠的多霍区。我们在开放式图表基准数据集(OGB)上进行的广泛实验表明,Neo-GNS(OGB)始终在链接预测中达到最先进的状态。我们的代码可在https://github.com/seongjunyun/Neo_GNNS上公开查阅。