Peer Disambiguation in Self-Reported Surveys using Graph Attention Networks

2025-03-25

AI Summary

Self-reported social network surveys suffer from pervasive link ambiguity—e.g., inconsistent interpretations of “friend”—distorting network topology and undermining peer intervention design and suicide risk prediction accuracy. To address this, we formally define and model bidirectional link ambiguity, comprising both existence judgment and candidate-link selection. We propose a graph attention network (GAT) framework that jointly leverages heterogeneous node attributes and structural topology, augmented with GNNExplainer for interpretable link discrimination. Evaluated on multiple real-world community datasets, our method significantly improves network reconstruction fidelity, boosting suicide risk prediction AUC by 8.2% while demonstrating robustness and generalizability. Our key contributions are: (1) establishing the first principled modeling paradigm for ambiguity in self-reported networks, and (2) uncovering socially grounded, interpretable relational patterns that drive predictive performance.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Studying peer relationships is crucial in solving complex challenges underserved communities face and designing interventions. The effectiveness of such peer-based interventions relies on accurate network data regarding individual attributes and social influences. However, these datasets are often collected through self-reported surveys, introducing ambiguities in network construction. These ambiguities make it challenging to fully utilize the network data to understand the issues and to design the best interventions. We propose and solve two variations of link ambiguities in such network data -- (i) which among the two candidate links exists, and (ii) if a candidate link exists. We design a Graph Attention Network (GAT) that accounts for personal attributes and network relationships on real-world data with real and simulated ambiguities. We also demonstrate that by resolving these ambiguities, we improve network accuracy, and in turn, improve suicide risk prediction. We also uncover patterns using GNNExplainer to provide additional insights into vital features and relationships. This research demonstrates the potential of Graph Neural Networks (GNN) to advance real-world network data analysis facilitating more effective peer interventions across various fields.

Problem

• Resolving link ambiguities in self-reported peer network data
• Improving network accuracy for better intervention design
• Enhancing suicide risk prediction through disambiguated networks

Innovation

• Uses Graph Attention Networks (GAT)
• Resolves link ambiguities in surveys
• Improves network accuracy and predictions