PhD Dissertation Defense at the College of Information Technology on Rumor Control Using Advanced Deep Learning Techniques

Duha Fadill Abbas
The Department of Software at the College of Information Technology, University of Babylon, witnessed the defense of the PhD dissertation submitted by Zainab Falah Hassan Shaker, under the supervision of Prof. Dr. Huda Naji Nawaf, entitled “Rumor Control Based on a Presumed Deep Dual Q Network Using Local and Global Centrality.” The defense was conducted in a rigorous academic setting that reflected the significance of the research in leveraging artificial intelligence and advanced deep reinforcement learning techniques to address the growing challenges associated with information diffusion in social networks.

The dissertation examined the critical issue of rumor propagation across social media platforms such as Facebook and Twitter, highlighting the societal risks posed by the rapid and uncontrolled spread of misleading information. The study emphasized that once rumors diffuse widely across a network, containing them becomes increasingly complex, thereby necessitating the development of intelligent and proactive control mechanisms capable of intervening at early stages of dissemination.

To address this challenge, the researcher proposed an integrated computational framework grounded in deep reinforcement learning, particularly Deep Q Network architectures, combined with structural analysis of social networks. The proposed approach exploits both local and global centrality measures to enhance the strategic selection of intervention nodes, while also incorporating community detection and diffusion modeling to better understand network topology and information flow dynamics.

The dissertation focused on constructing a comprehensive methodological model capable of capturing the structural and behavioral characteristics of information spread within digital communities. By integrating network structure awareness with adaptive decision-making mechanisms, the proposed framework aims to improve the efficiency of rumor containment strategies and contribute to the development of more resilient and reliable online environments.