HOPE improves graph node classification in networks with diverse connections
HOPE: Heterophily-Aware Open-Set Node Classification with Pseudo-Extrapolation
Machine Learning
Summary
Problems arise when trying to classify nodes in graphs where connected nodes do not share the same label, which is common in real-world networks. The authors propose a new method called HOPE that better handles these diverse connections by carefully filtering neighbors and creating synthetic examples to detect unknown classes. This approach improves the ability to identify nodes that belong to unknown categories and reduces confusion caused by mixed connections. Tests show HOPE performs better than existing methods on various datasets.
What this means in practice
- •For data engineers: Enhance anomaly detection in network data where connected entities have different labels due to heterophilic relationships.
- •For cybersecurity analysts: Improve identification of unknown threat types in network graphs with mixed connection patterns using more reliable unknown-class detection.
Authors
Yumeng Dai, Yue Tan, Yixin Liu, Chenxu Wang, Pinghui Wang, Tao Qin
Abstract
Standard open-set node classification methods rely on the homophily assumption, where connected nodes share labels. However, real-world graphs are often heterophilic, exposing the limitations of current methods and posing new challenges to open-set node classification. On the one hand, cross-class connectivity causes representations from different known or unknown classes to become intertwined after aggregation, undermining their discriminative capacity. On the other hand, structural mixture invalidates threshold-based open-set methods and cross-class feature interpolation, leading to unreliable unknown-class rejection. To address these challenges, we propose HOPE, a Heterophily-aware Open-set node classification method with Pseudo-Extrapolation. To adapt open-set graph neural networks (GNNs) to heterophilic scenarios, HOPE uses a structure-augmented feature initialization layer to capture multi-hop structural patterns. Meanwhile, we design a trustworthy neighborhood aggregation mechanism for standard GNNs to dynamically filter noisy cross-class neighbors. To enhance unknown-class rejection, we introduce a heterophily-guided pseudo-extrapolation strategy. It dynamically maintains known-class centers and extrapolates along cross-class neighborhood displacement directions, synthesizing pseudo-unknown proxies near structurally ambiguous regions. Finally, we optimize the network with joint classification and logit margin regularization, routing synthetic proxies into a dedicated rejection slot without imposing geometric margin constraints in the representation space. Extensive experiments on multiple datasets show that HOPE consistently outperforms state-of-the-art models, validating its effectiveness, robustness, and efficiency.