Adaptive authorization through transformer-based learning
Date
2025
Authors
Sinha, Pratik
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Access control is a cornerstone of information security, defining how entities interact with protected digital resources. Traditional rule-based frameworks, though effective in static environments, struggle to adapt to modern, data-intensive ecosystems where roles, attributes, and contextual conditions evolve continuously. Recent advances in machine learning have introduced new opportunities to automate access control through predictive and adaptive modeling yet progress remains constrained by the scarcity of real-world datasets, inconsistent benchmarking methodologies, and limited evaluation under controlled data conditions. This thesis presents a reproducible framework for evaluating machine-learning based access control models using synthetic, configurable datasets. The proposed data generation process emulates healthcare authorization structures, incorporating tunable role hierarchies, permission ratios, and anomaly patterns to simulate varying data noise and complexity. A suite of ML architectures, including decision-tree ensembles, feed-forward networks, residual networks, and transformer-based tabular models are systematically benchmarked using standardized preprocessing and evaluation metrics. Experimental results show that decision-tree ensembles provide strong baselines on small, structured datasets, while neural and transformer-based models generalize more reliably as data volume and complexity increase. This advantage depends on sufficiently rich, application-specific data where real datasets are typically sparse near decision boundaries, motivating targeted augmentation through synthetic generation around boundary cases. These findings validate the effectiveness of synthetic datasets for reproducible access-control research and demonstrate the impact of data scale on model elasticity and stability. Collectively, this work advances the development of reliable, ML-driven authorization systems through a transparent methodology for benchmarking and comparative analysis.
Description
Keywords
Access control, Machine learning, Synthetic data generation, Transformers