Drop the mask! GAMM-A Taxonomy for Graph Attributes Missing Mechanisms

2026-02-09

AI Summary

This work addresses the lack of a systematic classification for missing data mechanisms in attributed graphs, where conventional methods fail to capture the joint influence of graph structure and node attributes on missingness patterns. To this end, we propose the Graph-Aware Missingness Mechanism (GAMM) framework, which establishes the first missing data taxonomy tailored to attributed graphs by modeling missingness probabilities as a function of both node attributes and graph topology, thereby relaxing the i.i.d. assumption. Through comprehensive evaluation involving probabilistic modeling, graph neural networks, and imputation algorithms, our experiments demonstrate that state-of-the-art imputation methods suffer significant performance degradation under GAMM-defined graph-aware missingness scenarios, underscoring the practical relevance and challenge of the proposed classification framework.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Exploring missing data in attributed graphs introduces unique challenges beyond those found in tabular datasets. In this work, we extend the taxonomy for missing data mechanisms to attributed graphs by proposing GAMM (Graph Attributes Missing Mechanisms), a framework that systematically links missingness probability to both node attributes and the underlying graph structure. Our taxonomy enriches the conventional definitions of masking mechanisms by introducing graph-specific dependencies. We empirically demonstrate that state-of-the-art imputation methods, while effective on traditional masks, significantly struggle when confronted with these more realistic graph-aware missingness scenarios.

Problem

• attributed graphs
• missing data
• missingness mechanisms
• graph structure
• data imputation

Innovation

• Graph Attributes Missing Mechanisms
• missing data mechanisms
• attributed graphs
• graph-aware missingness
• data imputation