Year: 2025
Type: article
Abstract: The paper proposes a multimodal sentiment-based feedback system that can be employed to improve emotional awareness, engagement and creative performance in art education. With the combination of text sentiment analysis, speech emotion recognition, facial expression modeling, and digital behavior logging, the system is able to offer adaptive and emotionally appropriate feedback that helps students throughout intricate creative assignments. Implemented in digital painting, theatre, and music classes, the model was shown to achieve significant gains in the creativity scores, emotional strength, engagement rates, and the success rates of accomplishing the tasks compared to the conventional system of critique-based teaching. Quantitative approaches with the support of bar, line, boxplot, and heatmap visual tools prove the existence of strong correlations among emotional stability and artistic output, whereas qualitative intuitions point to more confident and less frustrated learners. The results reveal that sentiment-sensitive feedback does not only improve the performance, but the learning experience as a whole; empathetic and responsive as well as emotionally intelligent learning environments are created. The study represents a scalable framework and methodology proposal to implement affective computing into creative pedagogy, which should be applied in future educational systems based on AI.
Authors: Banashree Dash, Ananta Narayana, Akkamahadevi, Gourav Sood, Nishant Bhardwaj, Fariyah Saiyad
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W4417456495