{"id":22357,"date":"2026-02-19T16:29:24","date_gmt":"2026-02-19T13:29:24","guid":{"rendered":"https:\/\/su.edu.ye\/fcit\/?p=22357"},"modified":"2026-02-19T16:29:24","modified_gmt":"2026-02-19T13:29:24","slug":"phd-degree-awarded-to-mr-abdulsalam-mohammed-hussein-in-information-technology","status":"publish","type":"post","link":"https:\/\/su.edu.ye\/fcit\/en\/22357\/","title":{"rendered":"PhD Degree Awarded to Mr. Abdulsalam Mohammed Hussein in Information Technology"},"content":{"rendered":"

 <\/p>\n

Mr. Abdulsalam Mohammed Hussein Khaqo \u200e<\/strong>was awarded a PhD<\/strong> degree in Information Technology <\/strong>for his dissertation titled: Enhancing Asynchronous Advantage Actor-Critic Algorithm for Cybersecurity \u200eIntrusion Detection in Edge Computing\u200e<\/em><\/strong>, which was submitted to the Faculty of Computer and Information Technology \u2013 Sana\u2019a University. The dissertation defense was held on Thursday\u200e, February 12, 2026<\/strong>.<\/p>\n

The PhD Viva-Voce Committee, which was formed based on a resolution issued by the Graduate Studies and Scientific Research Council, consisted of the following:<\/p>\n

\u00a0<\/em><\/strong><\/p>\n\n\n\n\n\n\n
#<\/strong><\/td>\nCommittee Members<\/strong><\/td>\nDesignation<\/strong><\/td>\nPosition<\/strong><\/td>\n<\/tr>\n
1<\/td>\nAssoc. Prof. Naji Ali Abdullah Al-Shaibani \u200e<\/strong><\/td>\nInternal Examiner<\/td>\nChair<\/td>\n<\/tr>\n
2<\/td>\nProf. Sharaf Abdulhaq Mahyoub Al-Hamdi \u200e<\/strong><\/td>\nMain Supervisor<\/td>\nMember<\/td>\n<\/tr>\n
3<\/td>\nAssoc. Prof. Fouad Hassan Mohammed Yahya Abdulrazzaq \u200e<\/strong><\/td>\nExternal Examiner<\/td>\nMember<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

The dissertation aimed to:<\/p>\n

    \n
  • Develop adaptive class balancing using Dynamic Synthetic Minority Oversampling \u200ecombined with Edited Nearest Neighbors (DSMOTEENN) on the X-IIoTID \u200e\u200e<\/li>\n
  • Enhance the Worker component in the A3C algorithm to achieve faster convergence, \u200ereduced gradient variance, stable updates, and improved parameter \u200e\u200e<\/li>\n
  • Design the hybrid HADT algorithm integrating the enhanced A3C (suitable for \u200econtinuous-feature attack detection) with Decision Tree (DT) classifiers (effective \u200efor discrete-feature attacks).\u200e<\/li>\n<\/ul>\n

    The study yielded several key findings summarized as follows:<\/p>\n

      \n
    • The hybrid model effectively addressed data imbalance using DSMOTEENN, \u200eimproving rare-class detection and reducing error rates.\u200e<\/li>\n
    • The proposed enhancements to A3C within the EA3C model (including gradient \u200estabilization, weighted synchronization, and CNN\/FFNN-based feature \u200eextraction) achieved faster convergence, greater training stability, and superior \u200egeneralization compared to baseline models such as DT, DDQN, and AERL.\u200e<\/li>\n
    • The HADT algorithm achieved outstanding performance in Industrial IoT intrusion \u200edetection, with average Accuracy, Precision, Recall, and F1-Score exceeding \u200e\u200e8% in both binary and multi-class classification (up to 19 attack classes).\u200e<\/li>\n
    • ROC and AUC analyses showed near-perfect discrimination (0.9982\u20131.0000).\u200e<\/li>\n
    • The model demonstrated computational efficiency with reduced training time, \u200esuitable inference latency, and optimized energy consumption\u2014making it \u200eappropriate for resource-constrained edge computing environments.\u200e<\/li>\n
    • The proposed hybrid methodology outperformed traditional and standalone \u200ereinforcement learning approaches, establishing a new benchmark in Industrial IoT \u200ecybersecurity analytics.\u200e<\/li>\n<\/ul>\n

      In light of these findings, the researcher recommended the following:\u200e<\/p>\n

        \n
      • Expanding evaluation across diverse industrial datasets (energy, manufacturing, \u200ehealthcare) to validate generalizability.\u200e<\/li>\n
      • Integrating graph-based deep learning models (e.g., Graph Neural Networks and \u200eGraph Attention Networks) for detecting coordinated large-scale attacks.\u200e<\/li>\n
      • Exploring advanced reinforcement learning methods such as Proximal Policy \u200eOptimization (PPO) and Soft Actor-Critic (SAC).\u200e<\/li>\n
      • Incorporating Meta-Reinforcement Learning to enhance rapid adaptation to emerging \u200eattack patterns.\u200e<\/li>\n
      • Testing the system in real industrial environments with live traffic.\u200e<\/li>\n
      • Applying federated learning to enable secure model sharing across distributed edge \u200e\u200e<\/li>\n
      • Enhancing interpretability using Explainable AI (XAI) techniques such as SHAP and \u200e\u200e<\/li>\n
      • Applying model compression and quantization techniques for deployment on ultra-\u200econstrained embedded systems.\u200e<\/li>\n
      • Developing distributed Multi-Worker DRL architectures for cooperative defense in \u200elarge-scale IIoT networks.\u200e<\/li>\n
      • Moving toward Multi-Modal Intrusion Detection Systems integrating network flows, \u200edevice behavior, and sensor data.\u200e<\/li>\n<\/ul>\n

        The dissertation defense was attended by a number of academics, researchers, and specialists, students, colleagues, and the researcher\u2019s family.<\/p>\n

        \"\" \"\" \"\"<\/p>\n\n

        \n\n\t\t\n
          \n\t\t\t <\/ul>\n <\/div> \n","protected":false},"excerpt":{"rendered":"

            Mr. Abdulsalam Mohammed Hussein Khaqo \u200ewas awarded a PhD degree in Information Technology for his dissertation titled: Enhancing Asynchronous … <\/p>\n","protected":false},"author":128,"featured_media":22345,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_eb_attr":"","om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[121],"tags":[],"class_list":["post-22357","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-2"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t