|
December 20, 2024 - GTU Office of Press and Public Relations
Our academics' project proposal titled “Do Integrated Reports Published in Türkiye Signal Greenwashing? Deep Learning-Based Visual and Textual Analysis” is awarded support under the TÜBİTAK 3005 - Innovative Solutions in Social and Human Sciences Research Projects Support Program.
The project is led by Sibel Dinç-Aydemir (PhD), Rankings Coordinator at Gebze Technical University (GTU), with Assoc. Prof. Yakup Genç and Prof. Yusuf Sinan Akgül from GTU's Faculty of Engineering, Department of Computer Engineering as researchers. The project falls under the priority area of “scientific research on the impacts and risks of artificial intelligence technologies and systems on socio-economic structures” and aligns with the National Artificial Intelligence Strategy (UYZS).
This project focuses on integrated reports, which are a voluntary form of corporate reporting where organizations disclose both their financial activities and outputs and their environmental, social, and governance (ESG) practices, i.e., their sustainability activities. As such, there is no standardized template for these reports, and they vary in format and content. Unlike previous efforts, this project aims to perform a deep learning-based visual and textual analysis of these lengthy, content-rich documents, utilizing computer science techniques such as natural language processing (NLP) and computer vision. The project will benefit from both pre-trained and enhanced Large Language Models (LLMs). Furthermore, it will attempt to determine whether these reports signal greenwashing.
Offering an innovative, digital, and rich solution for analyzing integrated reports, the project is expected to provide the following benefits: organizations will be able to evaluate their own financial and sustainability reporting, regulatory and supervisory bodies will be able to monitor these reports, and third parties, particularly investors, will shape economic decisions based on these reports. Additionally, the interdisciplinary nature of this project is anticipated to open new research avenues for researchers in both fields involved. |
