An Interpretable and Accurate Deep-Learning

The deployment of automated deep-learning classifiers in clinical practice has the potential to streamline the diagnosis process and improve the diagnosis accuracy, but the acceptance of those classifiers relies on both their accuracy and interpretability. In general, accurate deep-learning classifiers provide little model interpretability, whi...

Research on Improving Cyber Resilience by Integrating the Zero Trust Security Model With the MITRE ATT&CK Matrix

With the advent of the digital information age, the dynamics of cyberspace are rapidly evolving, resulting in a significant increase in cyber threats. In this paper, we propose to integrate the Zero Trust (ZT) security model and the MITRE ATT&CK matrix to address the need for enhancing cyber resilience, which is an organization’s ability to recover...