AraMinds at AraHealthQA 2025: A Retrieval-Augmented Generation System for Fine-Grained Classification and Answer Generation of Arabic Mental Health Q&A
Year: 2025
Type: conference-paper
Abstract: We present a mental health support system for Arabic that can classify both patient questions and doctor answers, and generate answers for new questions.The classification model organizes the input text to understand better the intent of the user and the response style, while the generation model produces accurate and empathetic responses.In evaluations, our system ranked 3rd in answer classification and 4th in answer generation, with only a small margin from the top-ranked systems.These results highlight the effectiveness of multi-label classification and RAG for improving access to mental health information and support in Arabic.
Authors: Mohamed Zaytoon, Ahmed Mahmoud Salem, Ahmed Sakr, Hossam Elkordi
Institution: Alexandria University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4416037783