Predicting Properties of Nodes via Community-Aware Features
This work addresses node classification in network analysis, offering incremental improvements by leveraging community structure for feature design.
The paper tackles the problem of predicting node properties in networks by introducing community-aware node features, showing they contain unique information not captured by classical features or embeddings and improve node classification across synthetic and real-life networks.
This paper shows how information about the network's community structure can be used to define node features with high predictive power for classification tasks. To do so, we define a family of community-aware node features and investigate their properties. Those features are designed to ensure that they can be efficiently computed even for large graphs. We show that community-aware node features contain information that cannot be completely recovered by classical node features or node embeddings (both classical and structural) and bring value in node classification tasks. This is verified for various classification tasks on synthetic and real-life networks.