Scholarlink

<p>This research presents a novel Attribute-Based Access Control (ABAC) system that integrates "access history" as a key attribute, along with advanced deep learning techniques to improve decision-making in access control. The proposed system utilizes Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks to model time-series data, while Generative Adversarial Networks (GANs) generate synthetic data to address class imbalance issues. The architecture includes Policy Decision and Enforcement Points, real-time validation using an offline-trained model, and continuous learning from access logs. Extensive experiments on both real-world and synthetic datasets demonstrate that the proposed system achieves an accuracy of 98.5%, outperforming baseline models in handling complex access scenarios. The use of GANs enhances the model's robustness and adaptability, making it suitable for dynamic and evolving environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing time-series access control using deep recurrent neural networks and generative adversarial networks

  • 期刊论文

  • 发表日期：

    2025年1月13日

查看全文
  • Nasibeh Mohammadi,
  • Afshin Rezakhani,
  • Hamid Haj Seyyed Javadi,
  • Parvaneh Asghari

摘要

This research presents a novel Attribute-Based Access Control (ABAC) system that integrates "access history" as a key attribute, along with advanced deep learning techniques to improve decision-making in access control. The proposed system utilizes Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks to model time-series data, while Generative Adversarial Networks (GANs) generate synthetic data to address class imbalance issues. The architecture includes Policy Decision and Enforcement Points, real-time validation using an offline-trained model, and continuous learning from access logs. Extensive experiments on both real-world and synthetic datasets demonstrate that the proposed system achieves an accuracy of 98.5%, outperforming baseline models in handling complex access scenarios. The use of GANs enhances the model's robustness and adaptability, making it suitable for dynamic and evolving environments.