TEAChat: Promoting Educational Support for Autism with Conversational Artificial Intelligence

• Anderson Bispo dos Santos UPE
• Romário Elias da S. Santos UPE
• Carlo Marcelo R. da Silva UPE

DOI: https://doi.org/10.5753/wei.2025.9350

Abstract

This paper presents TEAChat, a tool based on generative artificial intelligence (AI) and a Retrieval-Augmented Generation (RAG) architecture, designed to support the education and inclusion of individuals with Autism Spectrum Disorder (ASD). The application personalizes responses according to user profiles (e.g., educators or family members), providing evidence-based information on diagnosis, therapies, and pedagogical strategies. The interface was developed using the Django framework and technologies such as large language models (LLMs), LangChain, and Gemini, offering an accessible and interactive experience. The results show TEAChat’s potential to raise awareness, reduce stigma, and serve as a practical support tool in educational, family, and clinical contexts.

References

Amazon Web Services (n.d.). O que é retrieval-augmented generation (rag)? [link]. Acessado em: 20 de novembro de 2024.

Andrade, A. M., Girardi, J. d. M., Melo, M. d. N., Pereira, D. C. R., Lima, M. d. G., Amorim, R. F. B. d., and Silva, W. M. C. d. (2024). Recursos educacionais para estudantes com transtorno do espectro do autismo (tea): SÍntese de evidÊncias qualitativas. Psicologia Escolar e Educacional, 28:e259575.

APA (2013). Diagnostic and Statistical Manual of Mental Disorders: DSM-5. American Psychiatric Publishing, Arlington, VA, 5 edition.

Bertacchini, F., Demarco, F., Scuro, C., Pantano, P., and Bilotta, E. (2023). A social robot connected with chatgpt to improve cognitive functioning in asd subjects. Frontiers in Psychology.

Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., et al. (2020). Language models are few-shot learners. arXiv pre-print arXiv:2005.14165. Large language models are massive neural networks trained on vast text corpora, enabling the generation and understanding of human-like text and performing tasks with minimal fine-tuning.

Curreri, V. and Fiora, A. (2023). Chatcare: A multimodal conversational interface for people with neurodevelopmental disorders.

Gurbuz, E., Hanley, M., and Riby, D. M. (2019). Estudantes universitários com autismo: As experiências sociais e acadêmicas da universidade no reino unido. Journal of Autism and Developmental Disorders, 49(2):617–631.

Hume, K., Steinbrenner, J. R., Odom, S. L., Morin, K. L., Nowell, S. W., Tomaszewski, B., Szendrey, S., McIntyre, N. S., Yücesoy-Özkan, S., and Savage, M. N. (2021). Evidence-based practices for children, youth, and young adults with autism: Third generation review. Journal of Autism and Developmental Disorders, 51(11):4013–4032. Erratum in: J Autism Dev Disord. 2023 Jan;53(1):514. DOI: 10.1007/s10803-022-05438-w.

Liu, F., Li, F., and Yang, Z. (2017). Chatbots: A new force in service industry. Journal of Service Theory and Practice, 27(3):642–655.

Liu, J. M., Li, D., Cao, H., Ren, T., Liao, Z., and Wu, J. (2023). Chatcounselor: A large language models for mental health support. In PGAI CIKM 2023, Birmingham, UK.

Narciso, R., da Silva Soares, L., Chiquera, D., Corrêa, L. L., das Chagas Oliveira, F. P., and da Silva, W. A. (2024). Conexões digitais no espectro autista: explorando potencialidades e promovendo inclusão. Revista Ibero-Americana de Humanidades, Ciências e Educação - REASE, 10(04):404–417.

Russell, S. J. and Norvig, P. (2021). Artificial Intelligence: A Modern Approach. Pearson, 4 edition.

Tuna, A. (2024). Use of chatbots to support the inclusion of people with autism spectrum disorder. In Integrating Generative AI in Education to Achieve Sustainable Development Goals.

UNESCO (2024). Guia para a IA generativa na educação e na pesquisa. UNESCO, Paris, França. Tradução para o português por Teresa Margarida Loureiro Cardoso e Viviane Cristina Marques.

USP (2023). Um retrato do autismo no Brasil. Acesso em: 20 out. 2024.