Two Researchers from the College of Information Technology Publish an Advanced Scientific Study on Movie Genre Classification Using Multimodal Models

Duhaa Fadill Abbas
As part of the ongoing efforts of the Department of Software Engineering at the College of Information Technology to promote rigorous scientific research and keep pace with rapid developments in artificial intelligence and machine learning, PhD candidate Yassin Khudair Abbas from the Department of Software Engineering, in collaboration with Dr. Ahmed Habib Al-Azzawi from the same department, has published a specialized scientific study addressing advanced techniques for multi-label movie genre classification based on intelligent multimodal models.

The study, entitled “A Proposed Multimodal Approach for Multi-Label Movie Genre Classification Based on a Transformer Model,” was published in the International Journal of Intelligent Engineering and Systems, an international peer-reviewed journal indexed in Scopus and ranked in the second quartile (Q2). The journal is published by The Intelligent Networks and Systems Society, reflecting the scientific value and academic significance of the research.

The paper presents an innovative model built upon Transformer architectures to address the challenge of multi-genre movie classification, a prominent issue in digital content recommendation systems, where a single movie may belong to multiple cinematic genres. The study integrates full textual data of movies with visual features extracted from movie posters, going beyond previous research that primarily relied on summaries or partial textual representations.

This research aims to develop more effective systems for content organization and personalization within large-scale movie platforms, thereby enhancing user experience and supporting intelligent recommendation systems. Furthermore, it contributes to expanding the scope of scientific research in the field of multimodal media classification.

Research link:https://doi.org/10.22266/ijies2026.0131.27