UFAZ Teacher’s Article on Artificial Intelligence Published in a Q1 Journal

The scientific article authored by Dr. Yusif Ibrahimli, a graduate of the French Azerbaijani University (UFAZ) and a lecturer in Computer Science, has been published on the IEEE platform, one of the world’s most reputable sources in the field of artificial intelligence, in the Q1-ranked journal IEEE Transactions on Artificial Intelligence.

The research paper titled “Explainable AI for Mental Disorder Detection on Social Media: A Survey and Outlook” provides a comprehensive review of existing approaches in the field of detecting mental disorders from social media data using artificial intelligence methods.

The article notes that social media platforms represent a rich source of data reflecting individuals’ emotional and psychological states. Such data creates significant opportunities for providing early-stage mental health support. However, the black-box nature of modern artificial intelligence models makes it difficult to understand how decisions are made. This, in turn, creates certain challenges in terms of clinical trust and practical application.

The study pays special attention to explainable artificial intelligence (XAI) approaches to address this issue. These include interpretable and understandable models, post hoc explanation methods, attention mechanisms, graph-based approaches, and explanation techniques based on large language models (LLMs), which have recently been widely used.

At the same time, the article also reviews the main databases and evaluation methods used in the field of social media-based mental disorder detection. The study highlights key challenges in this area, including data limitations, potential bias in models, and the difficulty of obtaining clinically meaningful explanations, and proposes future research directions.

You can access the article via the following link: https://ieeexplore.ieee.org/document/11495527