Oldie but Goodie: Re-illuminating Label Propagation on Graphs with Partially Observed Features

2025-08-02

AI Summary

Graph neural networks (GNNs) suffer significant performance degradation—or even underperform pure structural models—when node features are partially missing, particularly under low feature coverage. Method: This paper proposes GOODIE, a hybrid framework that revitalizes label propagation, designs a structure-feature attention mechanism for adaptive integration of topology and incomplete features, and incorporates pseudo-label contrastive learning to enhance embedding robustness. GOODIE unifies label propagation, feature propagation, and a GNN decoder within a single architecture, jointly optimizing structural modeling and feature imputation. Contribution/Results: GOODIE consistently outperforms state-of-the-art methods under both feature-scarce and feature-sufficient settings. Extensive experiments on multiple benchmark graph datasets validate its effectiveness, robustness, and generalizability across diverse missing-feature scenarios.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

In real-world graphs, we often encounter missing feature situations where a few or the majority of node features, e.g., sensitive information, are missed. In such scenarios, directly utilizing Graph Neural Networks (GNNs) would yield sub-optimal results in downstream tasks such as node classification. Despite the emergence of a few GNN-based methods attempting to mitigate its missing situation, when only a few features are available, they rather perform worse than traditional structure-based models. To this end, we propose a novel framework that further illuminates the potential of classical Label Propagation (Oldie), taking advantage of Feature Propagation, especially when only a partial feature is available. Now called by GOODIE, it takes a hybrid approach to obtain embeddings from the Label Propagation branch and Feature Propagation branch. To do so, we first design a GNN-based decoder that enables the Label Propagation branch to output hidden embeddings that align with those of the FP branch. Then, GOODIE automatically captures the significance of structure and feature information thanks to the newly designed Structure-Feature Attention. Followed by a novel Pseudo-Label contrastive learning that differentiates the contribution of each positive pair within pseudo-labels originating from the LP branch, GOODIE outputs the final prediction for the unlabeled nodes. Through extensive experiments, we demonstrate that our proposed model, GOODIE, outperforms the existing state-of-the-art methods not only when only a few features are available but also in abundantly available situations. Source code of GOODIE is available at: https://github.com/SukwonYun/GOODIE.

Problem

• Addresses missing node features in graph data
• Improves node classification with partial features
• Combines Label and Feature Propagation effectively

Innovation

• Hybrid Label and Feature Propagation approach
• Structure-Feature Attention for information significance
• Pseudo-Label contrastive learning for differentiation