Auto-ML Graph Neural Network Hypermodels for Outcome Prediction in Event-Sequence Data

2025-11-24

AI Summary

Accurate prediction of outcomes from complex event sequences remains challenging due to architectural uncertainty, hyperparameter sensitivity, and class imbalance. Method: This paper proposes HGNN(O), the first framework to deeply integrate AutoML with hierarchical graph neural network (HGNN) metamodels. It extends four GNN metamodel architectures, incorporates six GNN operators, and introduces a Bayesian optimization–driven adaptive tuning mechanism enabling end-to-end automatic architecture search, hyperparameter optimization, and pruning-based early stopping—without manual intervention or explicit class-imbalance handling. Results: On Traffic Fines and Patients datasets, HGNN(O) achieves accuracy >0.98 and weighted F1 = 0.86, respectively, demonstrating substantial improvements in predictive accuracy, robustness, and cross-domain generalization. Contribution: HGNN(O) establishes an automatically evolvable GNN metamodel paradigm, delivering an efficient, general-purpose AutoML solution for complex event sequence modeling.

Technology Category

• Machine Learning: Auto ML and Hyperparameter Tuning
• Search and Optimization: Metareasoning and Metaheuristics
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

This paper introduces HGNN(O), an AutoML GNN hypermodel framework for outcome prediction on event-sequence data. Building on our earlier work on graph convolutional network hypermodels, HGNN(O) extends four architectures-One Level, Two Level, Two Level Pseudo Embedding, and Two Level Embedding-across six canonical GNN operators. A self-tuning mechanism based on Bayesian optimization with pruning and early stopping enables efficient adaptation over architectures and hyperparameters without manual configuration. Empirical evaluation on both balanced and imbalanced event logs shows that HGNN(O) achieves accuracy exceeding 0.98 on the Traffic Fines dataset and weighted F1 scores up to 0.86 on the Patients dataset without explicit imbalance handling. These results demonstrate that the proposed AutoML-GNN approach provides a robust and generalizable benchmark for outcome prediction in complex event-sequence data.

Problem

• Predicting outcomes from complex event-sequence data automatically
• Automating architecture and hyperparameter selection for GNN models
• Handling balanced and imbalanced event logs without manual configuration

Innovation

• AutoML GNN hypermodel framework for outcome prediction
• Self-tuning Bayesian optimization with pruning mechanism
• Extends four architectures across six GNN operators