Feature Construction Using Network Control Theory and Rank Encoding for Graph Machine Learning

2025-07-20

AI Summary

To address the performance degradation of Graph Neural Networks (GNNs) caused by missing node attributes in social network classification, this paper proposes a featureless node representation method grounded in network control theory. The core method introduces average controllability—a topology-driven centrality measure—into graph representation learning for the first time, and designs a ranking-based encoding strategy to map it into fixed-dimensional, learnable dense feature vectors. Critically, the approach requires no raw node attributes, relying solely on graph structure to generate highly discriminative features. Experiments on four real-world social network datasets demonstrate that, when integrated with GraphSAGE, the method achieves a ROC AUC of 73.9%, outperforming one-hot encoding by 5.2 percentage points and substantially surpassing conventional centrality-based baselines. This work establishes a novel paradigm for attribute-free graph learning.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Computer Vision: Representation Learning for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Social Networks and Social Media: Computational social science
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In this article, we utilize the concept of average controllability in graphs, along with a novel rank encoding method, to enhance the performance of Graph Neural Networks (GNNs) in social network classification tasks. GNNs have proven highly effective in various network-based learning applications and require some form of node features to function. However, their performance is heavily influenced by the expressiveness of these features. In social networks, node features are often unavailable due to privacy constraints or the absence of inherent attributes, making it challenging for GNNs to achieve optimal performance. To address this limitation, we propose two strategies for constructing expressive node features. First, we introduce average controllability along with other centrality metrics (denoted as NCT-EFA) as node-level metrics that capture critical aspects of network topology. Building on this, we develop a rank encoding method that transforms average controllability or any other graph-theoretic metric into a fixed-dimensional feature space, thereby improving feature representation. We conduct extensive numerical evaluations using six benchmark GNN models across four social network datasets to compare different node feature construction methods. Our results demonstrate that incorporating average controllability into the feature space significantly improves GNN performance. Moreover, the proposed rank encoding method outperforms traditional one-hot degree encoding, improving the ROC AUC from 68.7% to 73.9% using GraphSAGE on the GitHub Stargazers dataset, underscoring its effectiveness in generating expressive and efficient node representations.

Problem

• Enhancing GNN performance in social network classification tasks
• Addressing lack of node features in social networks
• Improving feature representation with rank encoding method

Innovation

• Utilizes average controllability for node features
• Introduces rank encoding for fixed-dimensional features
• Combines NCT-EFA metrics to enhance GNN performance