A Detailed Seminar of Ph.D. Student Yaseen Khudair Abbas Abd in the Department of Software

Duhaa Fadill Abbas
As part of the ongoing scientific activities held by the Department of Software at the College of Information Technology – University of Babylon, Ph.D. student Yaseen Khudair Abbas Abd delivered a A Detailed Seminar to present and discuss his dissertation entitled:
"A Proposed Automated Approach for Movie Genre Classification and Age Appropriateness Based on Scripts,"
under the supervision of Asst. Prof. Dr. Ahmed Habeeb Saeed, and in the presence of the department’s scientific committee members.

The study aims to develop a unified intelligent model capable of classifying movies by genre and determining their age appropriateness, relying on the analysis of textual and visual content. This approach comes in response to the increasing demand for accurate technological tools that help regulate cinematic content in alignment with cultural and ethical standards.

The proposed methodology is structured into multiple stages. The first stage focuses on movie genre classification using a multimodal framework that integrates textual features extracted via the RoBERTa model—enhanced through a chunking mechanism to process full-length scripts—and visual features obtained from movie posters using the EfficientNet model. To address challenges related to multi-label classification and class imbalance, an asymmetric loss function was utilized during training.

The second stage addresses age appropriateness classification based on the MPAA rating system, leveraging sentiment analysis derived from the movie scripts. To enhance the performance in imbalanced datasets, the Adaptive Synthetic Sampling (ADASYN) technique was employed for data balancing