PROTEA-M: FRAMEWORK MULTIMODAL PARA COMUNICAÇÃO INCLUSIVA EM SITUAÇÕES DE INUNDAÇÃO
Resumo
Os sistemas de alerta precoce constituem a principal ferramenta para notificação e orientação da população em situações de emergência, porém ainda utilizam mensagens padronizadas que não consideram diferenças cognitivas e comunicativas. Essa limitação impacta de forma direta pessoas com Transtorno do Espectro Autista, que demandam formas estruturadas e previsíveis de organização da informação para compreensão e tomada de decisão em contextos de risco. Este trabalho apresenta o framework PROTEA-M (Program for Readiness and Organized Training in Emergencies for Autistic Individuals - Multimodal), um modelo preditivo-generativo multimodal voltado à produção de conteúdos inclusivos para preparação em cenários de inundação. O modelo estima o perfil comunicativo e o plano de intervenção e, a partir dessa estimativa, gera Histórias Sociais personalizadas. O módulo preditivo utiliza ensemble de Extreme Learning Machines com fusão de evidências e calibração probabilística, apresentando desempenho superior aos baselines, com acurácia de 0,6923 e ROC-AUC de 0,9782, além de melhor qualidade probabilística. O módulo generativo produz narrativas multimodais por meio de modelos de linguagem, traduzindo variáveis clínicas em sequências visuais e textuais organizadas. O PROTEA-M contribui para a evolução dos sistemas de alerta ao incorporar personalização comunicacional e acessibilidade cognitiva.
Biografia do Autor
Desenvolvedor mobile Flutter, especializado em design responsivo e integração de aplicações utilizando API REST. Bacharel em Engenharia da Computação pela Universidade Paulista e mestre em Engenharia da Computação pela Universidade Pernambuco com certificação como Professional Scrum Master pela scrum.org e como Flutter Developer pela AndroidATC.
Graduado em Engenharia Elétrica pela Universidade Federal de Pernambuco (1978), mestre em Engenharia Elétrica pela Universidade Federal de Pernambuco (1995) e doutor em Ciências da Computação pelo Centro de Informática da Universidade Federal de Pernambuco (2007). Adjunto II da Universidade Católica de Pernambuco e da Escola Politécnica da Universidade de Pernambuco.
Referências
ABDULLAH, Maria et al. Artificial intelligence-based assistive technologies for autism spectrum disorder: a systematic review. Journal of Autism and Developmental Disorders, v. 52, n. 9, p. 4029-4050, 2022.
AHMED, S. et al. An optimized Kernel Extreme Learning Machine for the classification of the autism spectrum disorder by using gaze tracking images. Applied Soft Computing, v. 120, p. 108654, 2022. DOI: https://doi.org/10.1016/j.asoc.2022.108654
BALOG-WAY, Dominic H. P.; MCCOMAS, Katherine A.; BESLEY, John C. The evolving field of risk communication. Risk Analysis, v. 40, n. S1, p. 2240-2262, 2020. DOI: https://doi.org/10.1111/risa.13615
BASHER, Reid. Global early warning systems for natural hazards: systematic and people-centred. Philosophical Transactions of the Royal Society A, v. 364, n. 1845, p. 2167-2182, 2006. DOI: https://doi.org/10.1098/rsta.2006.1819
BEUKELMAN, David R.; MIRENDA, Pat. Augmentative and alternative communication: supporting children and adults with complex communication needs. 4. ed. Baltimore: Paul H. Brookes Publishing, 2013.
BOGDASHINA, Olga. Sensory perceptual issues in autism and Asperger syndrome: different sensory experiences, different perceptual worlds. 2. ed. London: Jessica Kingsley Publishers, 2016.
BREIMAN, Leo. Random forests. Machine Learning, v. 45, n. 1, p. 5-32, 2001. DOI: https://doi.org/10.1023/A:1010933404324
BRIER, Glenn W. Verification of forecasts expressed in terms of probability. Monthly Weather Review, v. 78, n. 1, p. 1-3, 1950. DOI: https://doi.org/10.1175/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2
BROWN, Tom et al. Language models are few-shot learners. Advances in Neural Information Processing Systems, v. 33, p. 1877-1901, 2020.
CENTERS FOR DISEASE CONTROL AND PREVENTION (CDC). Data and statistics on autism spectrum disorder. Atlanta: CDC, 2023. Disponível em: https://www.cdc.gov/autism/data-research/index.html Acesso em: 25 maio 2026.
CERDA, Patricio; VAROQUAUX, Gaël; KÉGL, Balázs. Similarity encoding for learning with dirty categorical variables. Machine Learning, v. 107, p. 1477-1494, 2018. DOI: https://doi.org/10.1007/s10994-018-5724-2
COHEN, Jacob. A coefficient of agreement for nominal scales. Educational and Psychological Measurement, v. 20, n. 1, p. 37-46, 1960. DOI: https://doi.org/10.1177/001316446002000104
CORTES, Corinna; VAPNIK, Vladimir. Support-vector networks. Machine Learning, v. 20, p. 273-297, 1995. DOI: https://doi.org/10.1007/BF00994018
COVER, Thomas; HART, Peter. Nearest neighbor pattern classification. IEEE Transactions on Information Theory, v. 13, n. 1, p. 21-27, 1967. DOI: https://doi.org/10.1109/TIT.1967.1053964
DIETTERICH, Thomas G. Ensemble methods in machine learning. In: MULTIPLE CLASSIFIER SYSTEMS. Berlin: Springer, 2000. p. 1-15. DOI: https://doi.org/10.1007/3-540-45014-9_1
ELDAWLATLY, Abdelazeem. Writing the methods section. Saudi Journal of Anaesthesia, v. 13, supl. 1, p. S10-S11, 2019. DOI: https://doi.org/10.4103/sja.SJA_685_18
FAWCETT, Tom. An introduction to ROC analysis. Pattern Recognition Letters, v. 27, n. 8, p. 861-874, 2006. DOI: https://doi.org/10.1016/j.patrec.2005.10.010
FAYYAD, Usama; PIATETSKY-SHAPIRO, Gregory; SMYTH, Padhraic. From data mining to knowledge discovery in databases. AI Magazine, v. 17, n. 3, p. 37-54, 1996. DOI: https://doi.org/10.1609/aimag.v17i3.1230
FENG, Yi et al. SS-GEN: a social story generation framework with large language models. In: Proceedings of the AAAI Conference on Artificial Intelligence. v. 39, n. 2, p. 1300-1308, 2025.
FRITH, Uta. Autism: explaining the enigma. Oxford: Blackwell, 2003.
GRAY, Carol. The new social story book. 10. ed. Arlington: Future Horizons, 2010.
GUO, Chuan et al. On calibration of modern neural networks. In: International Conference on Machine Learning. [S.l.]: PMLR, 2017. p. 1321-1330.
HASTIE, Trevor; TIBSHIRANI, Robert; FRIEDMAN, Jerome. The elements of statistical learning. 2. ed. New York: Springer, 2009.
HOERL, Arthur E.; KENNARD, Robert W. Ridge regression: biased estimation for nonorthogonal problems. Technometrics, v. 12, n. 1, p. 55-67, 1970. DOI: https://doi.org/10.1080/00401706.1970.10488634
HUANG, Guang-Bin; ZHU, Qin-Yu; SIEW, Chee-Kheong. Extreme learning machine: theory and applications. Neurocomputing, v. 70, n. 1-3, p. 489-501, 2006. DOI: https://doi.org/10.1016/j.neucom.2005.12.126
INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (IBGE). Censo Demográfico 2022. Rio de Janeiro: IBGE, 2022. Disponível em: https://www.ibge.gov.br Acesso em: 25 maio 2026.
INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE (IPCC). Climate change 2023: synthesis report. Geneva: IPCC, 2023. Disponível em: https://www.ipcc.ch/report/ar6/syr/ Acesso em: 25 maio 2026.
KENT, Rachel et al. Preparing for emergencies among families of children with autism spectrum disorder. Journal of Autism and Developmental Disorders, v. 43, n. 3, p. 666-675, 2013. DOI: https://doi.org/10.1007/s10803-012-1618-0
KOHAVI, Ron. A study of cross-validation and bootstrap for accuracy estimation and model selection. In: International Joint Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann, 1995. p. 1137-1145.
KOHONEN, Teuvo. Self-organizing maps. 3. ed. Berlin: Springer, 2001.
KOKINA, Anastasia; KERN, Lee. Social story interventions for students with autism spectrum disorders: a meta-analysis. Journal of Autism and Developmental Disorders, v. 40, n. 7, p. 812-826, 2010. DOI: https://doi.org/10.1007/s10803-009-0931-0
LINDELL, Michael K.; PERRY, Ronald W. The protective action decision model: theoretical modifications and additional evidence. Risk Analysis, v. 32, n. 4, p. 616-632, 2012. DOI: https://doi.org/10.1111/j.1539-6924.2011.01647.x
LUNDGREN, Regina E.; MCMAKIN, Andrea H. Risk communication: a handbook for communicating environmental, safety, and health risks. 6. ed. Hoboken: Wiley, 2018.
LUNDBERG, Scott M.; LEE, Su-In. A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems, 2017.
MICCI-BARRECA, Daniele. A preprocessing scheme for high-cardinality categorical attributes in classification and prediction problems. ACM SIGKDD Explorations Newsletter, v. 3, n. 1, p. 27-32, 2001. DOI: https://doi.org/10.1145/507533.507538
MILETI, Dennis S.; SORENSEN, John H. Communication of emergency public warnings. Oak Ridge: ORNL, 1990.
NAEINI, Mahdi Pakdaman; COOPER, Gregory; HAUSKRECHT, Milos. Obtaining well calibrated probabilities using Bayesian binning. In: AAAI Conference on Artificial Intelligence. 2015. p. 2901-2907.
PEEK, Lori; STOUGH, Laura. Children with disabilities in the context of disaster: a social vulnerability perspective. Child Development, v. 81, n. 4, p. 1260-1270, 2010. DOI: https://doi.org/10.1111/j.1467-8624.2010.01466.x
QUINLAN, J. Ross. Induction of decision trees. Machine Learning, v. 1, p. 81-106, 1986. DOI: https://doi.org/10.1007/BF00116251
REIMERS, Nils; GUREVYCH, Iryna. Sentence-BERT: sentence embeddings using Siamese BERT-networks. In: Conference on Empirical Methods in Natural Language Processing. Hong Kong: ACL, 2019. p. 3982-3992. DOI: https://doi.org/10.18653/v1/D19-1410
ROBERTSON, Caroline E.; BARON-COHEN, Simon. Sensory perception in autism. Nature Reviews Neuroscience, v. 18, n. 11, p. 671-684, 2017. DOI: https://doi.org/10.1038/nrn.2017.112
ROWLAND, Charity. Matriz de Comunicação. Portland: Oregon Health & Science University, 2011.
RUTHERFORD, Marion et al. Visual supports at home and in the community for autistic children: a scoping review. Autism, v. 24, n. 7, p. 1737-1752, 2020. DOI: https://doi.org/10.1177/1362361320901572
SAITO, Takaya; REHMSMEIER, Marc. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLOS ONE, v. 10, n. 3, e0118432, 2015. DOI: https://doi.org/10.1371/journal.pone.0118432
SHAH, Ashfaq Ahmad et al. Community social barriers to non-technical aspects of flood early warning systems and NGO-led interventions: the case of Pakistan. Frontiers in Earth Science, v. 11, p. 1068721, 2023. DOI: https://doi.org/10.3389/feart.2023.1068721
STOUGH, Laura M.; KELMAN, Ilan. People with disabilities and disasters. New York: Palgrave Macmillan, 2018.
SUTTON, Jeannette; TIERNEY, Kathleen. Disaster preparedness: concepts, guidance, and research. Boulder: University of Colorado, 2006.
TEST, David W. et al. Evidence-based practices for individuals with autism. Exceptional Children, v. 78, n. 2, p. 135-155, 2011. DOI: https://doi.org/10.1177/001440291107800201
UNITED NATIONS CHILDREN'S FUND (UNICEF). Disability-inclusive disaster risk reduction. New York: UNICEF, 2023.
UNITED NATIONS OFFICE FOR DISASTER RISK REDUCTION (UNDRR). Global assessment report on disaster risk reduction 2022. Geneva: UN, 2022.
UNITED NATIONS OFFICE FOR DISASTER RISK REDUCTION (UNDRR). Sendai framework for disaster risk reduction 2015-2030. Geneva: UN, 2015.
VARMA, Sudhir; SIMON, Richard. Bias in error estimation when using cross-validation for model selection. BMC Bioinformatics, v. 7, n. 91, 2006. DOI: https://doi.org/10.1186/1471-2105-7-91
VASWANI, Ashish et al. Attention is all you need. Advances in Neural Information Processing Systems, v. 30, 2017.
WORLD BANK. Inclusive early warning systems: a lifeline for all. Washington: World Bank, 2022.
WORLD METEOROLOGICAL ORGANIZATION (WMO). State of the global climate 2021. Geneva: WMO, 2021.
Licença
Copyright (c) 2026 RECIMA21 - Revista Científica Multidisciplinar - ISSN 2675-6218
Este trabalho está licenciado sob uma licença Creative Commons Attribution 4.0 International License.
Os direitos autorais dos artigos/resenhas/TCCs publicados pertecem à revista RECIMA21, e seguem o padrão Creative Commons (CC BY 4.0), permitindo a cópia ou reprodução, desde que cite a fonte e respeite os direitos dos autores e contenham menção aos mesmos nos créditos. Toda e qualquer obra publicada na revista, seu conteúdo é de responsabilidade dos autores, cabendo a RECIMA21 apenas ser o veículo de divulgação, seguindo os padrões nacionais e internacionais de publicação.