NSF Award
2427894
This research project focuses on improving data privacy and security within complex networks, a key issue in today's technology-driven society. As technology advances, stronger protections against new threats become crucial for complex systems, such as infrastructure networks, the Internet of Things, and global trade systems. The project outlines a comprehensive strategy to enhance network security: introducing preventive measures to stop data breaches before they happen, setting up advanced monitoring for quick detection of irregularities, and developing specific strategies for responding to security breaches. The project's implications extend far beyond security, poised to influence fields as diverse as social sciences, biological network analysis, and misinformation studies. Importantly, the project is also committed to educational excellence and knowledge dissemination. Integrating the project's findings into the academic curriculum aims to bridge the gap between theoretical research and practical application, nurturing a new generation of experts skilled in the nuances of network security. The project also extends significantly into student development, aiming to foster a nurturing environment for student involvement. It provides valuable research opportunities for both undergraduates and graduates. Through hands-on projects, software development, and data analysis tasks, students will gain practical experience and insights into real-world applications of their studies. Special emphasis is also placed on the involvement of students from traditionally underrepresented groups, which seeks to cultivate a diverse and vibrant research community. Furthermore, the initiative plans to release a suite of open-source software tools and databases, democratizing access to state-of-the-art methods in network analysis and security. These resources, coupled with workshops and tutorials, will empower researchers, practitioners, and policymakers to implement effective security strategies, fostering a safer digital ecosystem for all. Through these multifaceted efforts, the project contributes to the scientific understanding of network protection and champions the cause of equity and inclusiveness, ensuring that the benefits of secure and resilient networks are accessible to a broad swath of society. <br/><br/>The project stands out for its innovative integration of statistical modeling, data privacy, deep learning, and optimization techniques, aimed at addressing the multifaceted challenges of securing complex networks. This blend of approaches, unusual in its breadth and depth, sets the endeavor apart in complex network research. First, the project introduces a cutting-edge approach to protect network data against privacy breaches and adversarial attacks, focusing on maintaining data utility. It establishes a novel framework for latent node-level differential privacy, applying distribution-invariant mechanisms to ensure that released network data safeguards both privacy and security without losing its intrinsic value. Second, the project develops flexible yet robust methodologies for timely detection and localization of network anomalies, utilizing nonparametric estimation for dynamic network behavior modeling. It aims to enhance network security monitoring by accurately identifying non-stationary change points and pinpointing anomalies, enabling prompt and effective responses to emerging threats. Third, the project introduces a comprehensive approach to tracking down the sources of misinformation in social networks, utilizing a broad spectrum of methods, including spectral methods, graphical models, and graph neural networks. It aims to efficiently identify misinformation origins and devise strategies to curtail its spread, thereby enhancing the reliability and integrity of information across social platforms.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
