Evaluating Student Emotional Well-Being in Humanitarian Educational Contexts Using an AI-Based Sentiment Analysis System

This study explores the challenge of monitoring student emotional well-being in humanitarian educational contexts, where learners are often exposed to stress, anxiety, and trauma that negatively impact their learning and mental health. Early detection of these emotional issues remains difficult due to limited resources and large class sizes. To address this problem, an AI-based system utilizing sentiment analysis is proposed to automatically identify students’ emotional states from textual data. The system employs natural language processing and transformer-based deep learning models, such as BERT and RoBERTa, and supports multilingual inputs. It integrates components including emotion classification, real-time alerts, and educator guidance to enable timely interventions. The results indicate that the proposed approach improves the accuracy of emotion detection and enhances the ability of educators to identify at-risk students. These findings are consistent with previous studies, where transformer-based models achieved high performance in sentiment analysis tasks, with accuracy levels exceeding 90% in certain contexts. The system also demonstrated effective real-time monitoring capabilities, contributing to early detection and response. Overall, this approach supports the development of more responsive educational environments and highlights the potential of AI-driven solutions in improving student well-being in humanitarian settings. This work was conducted at Arab International University (AIU), Syria. The official website of the university is: https://www.aiu.edu.sy