Software Department Discusses a Pre-Writing Seminar for a Ph.D. Student on Fake News Detection Using Deep Learning Attention Mechanisms

Duhaa Fadill Abbas
As part of its ongoing academic activities, the Software Department at the College of Information Technology – University of Babylon organized a pre-writing seminar for Ph.D. student Iman Qais Abdul-Jalil, under the supervision of Prof. Israa Hadi Ali. During the seminar, the student delivered a detailed presentation of her dissertation proposal entitled:
“Detecting Fake News on Social Networks Based on Textual and Visual Sources Using the Deep Learning Attention Mechanism.”

In her presentation, the student emphasized the growing significance of this research topic, particularly in light of the increasing popularity of social media platforms, which has been accompanied by an escalation in misinformation campaigns, a decline in trust toward disseminated content, and the difficulty of verifying information accuracy through human capabilities alone. She highlighted that most existing fake news detection systems concentrate primarily on the textual dimension of news items while neglecting the substantial role played by accompanying visual content, which often serves as the initial driver for reading and subsequent sharing.

The student further explained that this reliance on textual analysis alone reduces the accuracy of detecting fake news that includes images, underscoring the necessity of developing a more effective multimodal automated model. She noted that the primary objective of her research is to design an integrated framework based on deep learning and the attention mechanism to enable rapid and accurate detection of fake news by processing both textual and visual data within a carefully curated dataset.

The seminar concluded with valuable feedback and recommendations from the department’s scientific committee, aimed at refining the research framework and enhancing the quality and originality of the expected outcomes.