A scientific study co-authored by Lecturer Özge Demir from the Big Data Analytics Program of Beykoz University Vocational School has been published in Scientific Reports, an international academic journal.
In the study, single- and dual-backbone deep learning architectures were comparatively analyzed, and explainability methods were integrated to enhance the interpretability of model decisions. The findings reveal the impact of different architectural structures on performance, while also contributing to the development of more reliable and interpretable artificial intelligence systems.
Scientific Reports is a multidisciplinary journal within the Nature Portfolio and is indexed in the SCI-Expanded database under Web of Science, holding a significant position in the international academic community. With an impact factor of approximately 3.8, the journal provides broad accessibility and global visibility for published research.
This publication stands out as an important achievement that strengthens the research-oriented approach and international academic visibility of Beykoz University’s Big Data Analytics Program.
We congratulate our lecturer and wish her continued success.
This content was updated on 22/04/2026.