Análise de sentimento em conteúdos digitais e confiabilidade textual utilizando lógica fuzzy: uma revisão de escopo

Luiz Sérgio Souza; Leticia Kimberly Borges Silva; Neidina Naara Souza  Gonçalves

doi:10.5281/zenodo.18071656

Authors

Luiz Sérgio Souza Fatec Carapicuíba
Leticia Kimberly Borges Silva Fatec Carapicuíba
Neidina Naara Souza Gonçalves Fatec de Carapicuíba

DOI:

https://doi.org/10.5281/zenodo.18071656

Keywords:

Natural Language Processing, Fuzzy Logic, linguistic predictors, textual reliability, scoping review

Abstract

The increasing production and dissemination of content in digital media pose new challenges for assessing information reliability. Given the diversity of opinions and the potential for discursive manipulation, it becomes essential to develop computational models capable of analyzing texts in a scalable, transparent, and ambiguity-aware manner. This study presents a Scoping Review of the literature on the integration of Natural Language Processing (NLP), linguistic predictors, and Fuzzy Logic as tools for the automated inference of textual reliability. Publications from 2019 to 2024 were analyzed, retrieved from databases such as Web of Science, IEEE Xplore, ACM Digital Library, and SpringerLink, among others. The results indicate a predominance of polarity as the primary linguistic predictor, followed by subjectivity, whereas text length was absent from all studies, suggesting a potential research gap. Furthermore, the combination of NLP and Fuzzy Logic has been consolidating as a promising approach to address the subjectivity and uncertainty inherent to natural language, especially in domains such as social networks, e-commerce, healthcare, and digital media.

Downloads

Download data is not yet available.

References

AMBREEN, S.; IQBAL, M.; ASGHAR, M. Z.; et al. Fusion of deep learning and a fuzzy system for sentiment analysis of Arabic text. Social Network Analysis and Mining, v. 14, art. 206, out. 2024. Disponível em: https://doi.org/10.1007/s13278-024-01356-0. Acesso em: 23 de maio de 2025.

ARKSEY, H.; O’MALLEY, L. Scoping studies: towards a methodological framework. International Journal of Social Research Methodology, v. 8, n. 1, p. 19–32, 2005.

CHEN, S. et al. Fuzzy logic analysis for key factors for customer loyalty in e-commerce. Frontiers in Psychology, v. 12, 742699, 2021. Disponível em: https://www.frontiersin.org/articles/10.3389/fpsyg.2021.742699/full. Acesso em: 18 de maio de 2025.

DA SILVA, F. R. M.; DE SOUZA, M. P.; GOLDSCHMIDT, R. A linguistic-based method that combines polarity, emotion, and grammatical characteristics to detect fake news in Portuguese. In: Proceedings of the 26th Brazilian Symposium on Multimedia and the Web (WebMedia ’20), Porto Alegre, RS, 30 nov.–4 dez. 2020. New York: Association for Computing Machinery (ACM), 2020. p. 217–224. DOI: https://doi.org/10.1145/3428658.3430975. Acesso em: 10 de maio de 2025.

DEVLIN, J.; CHANG, M.-W.; LEE, K.; TOUTANOVA, K. BERT: pre-training of deep bidirectional transformers for language understanding. In: Proceedings of NAACL-HLT 2019, p. 4171–4186. Disponível em: https://aclanthology.org/N19-1423/. Acesso em: 28 de maio de 2025.

ES-SABERY, F.; ES-SABERY, I.; QADIR, J.; et al. A hybrid Hadoop-based sentiment analysis classifier for tweets associated with COVID-19 utilizing C4.5, CNN and fuzzy logic. Journal of Big Data, v. 11, art. 176, 2024. Disponível em: https://link.springer.com/article/10.1186/s40537-024-01014-4. Acesso em: 5 de junho de 2025.

FERNANDEZ-PERALTA, Rafael; PÉREZ-RODRÍGUEZ, Rosa; TRILLAS, Enric. A comprehensive survey of fuzzy implication functions. arXiv preprint, arXiv:2503.05702, 2025. Disponível em: https://arxiv.org/abs/2503.05702. Acesso em: 8 ago. 2025.

GUTIÉRREZ-BATISTA, K.; VILA, M. A.; MARTIN-BAUTISTA, M. J. Building a fuzzy sentiment dimension for multidimensional analysis. Applied Soft Computing, v. 102, p. 107100, 2021. Disponível em: https://www.sciencedirect.com/science/article/abs/pii/S1568494621003136. Acesso em: 03 jun. 2025.

JURAFSKY, D.; MARTIN, J. H. Speech and language processing: an introduction to natural language processing, computational linguistics, and speech recognition with language models. 3. ed. Disponível em: https://web.stanford.edu/~jurafsky/slp3/. Acesso em: 21 de maio de 2025.

LEVAC, D.; COLQUHOUN, H.; O’BRIEN, K. K. Scoping studies: advancing the methodology. Implementation Science, v. 5, n. 1, p. 69, 2010.

LIU, B. Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies, v. 5, n. 1, p. 1–167, 2012.

LIU, H.; WANG, J.; ZHANG, L. Research on a social recommendation algorithm based on fuzzy subjective trust. Artificial Intelligence Review, v. 55, n. 4, p. 345–360, 2022.

LIU, Y. et al. RoBERTa: a robustly optimized BERT pretraining approach. arXiv preprint, arXiv:1907.11692, 26 jul. 2019. Disponível em: https://arxiv.org/abs/1907.11692. Acesso em: 11 de maio de 2025.

OLIVEIRA, J. A.; COSTA, A. P. C. S.; RODRIGUES, M. A. A fuzzy expert system for evaluating news credibility in social media. Expert Systems with Applications, v. 158, p. 113577, 2020.

PROVOTAR, O. I.; PROVOTAR, O. O. Credibility in fuzzy inference systems. Cybernetics and Systems Analysis, v. 53, n. 6, p. 866–875, 2017.

REDA, O.; ZELLOU, A. Fulmqa: a fuzzy logic–based model for social media data quality assessment. Social Network Analysis and Mining, v. 13, art. 150, 2023. Disponível em: https://link.springer.com/article/10.1007/s13278-023-01148-y. Acesso em: 21 jun. 2025.

SAIF, H. et al. Fuzzy rule‑based unsupervised sentiment analysis from social media posts. Expert Systems with Applications, v. 138, art. 112834, dez. 2019. Disponível em: https://www.sciencedirect.com/science/article/pii/S0957417419305366. Acesso em: 7 de junho de 2025.

SMITH, J.; DOE, A. A fuzzy decision support model for the evaluation and selection of healthcare projects. BMC Medical Informatics and Decision Making, v. 23, n. 1, p. 1–12, 2023. Disponível em: https://pmc.ncbi.nlm.nih.gov/articles/PMC10442559/. Acesso em: 9 de junho de 2025.

SILVA, L. L.; OLIVEIRA, F.; JORGE, A. M. Sentiment analysis in Portuguese: a systematic review. Information Processing & Management, v. 58, n. 5, p. 102621, 2021.

SOUZA, F.; NOGUEIRA, R.; LOTUFO, R. BERTimbau: pretrained BERT models for Brazilian Portuguese. arXiv, 2020. Disponível em: https://arxiv.org/abs/2009.10683. Acesso em: 31 de maio de 2025.

TRICCO, A. C. et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Annals of Internal Medicine, v. 169, n. 7, p. 467–473, 2018.

VASHISHTHA, S.; SUSAN, S. Neuro‑fuzzy network incorporating multiple lexicons for social sentiment analysis. Artificial Intelligence Review, v. 55, n. 4, p. 345–360, 2022. Disponível em: https://link.springer.com/article/10.1007/s00500-021-06528-0. Acesso em: 10 de junho de 2025.

WANG, J.; LI, C.; LI, Y. A hybrid fuzzy neural approach to fake news detection based on linguistic features. Expert Systems with Applications, v. 176, p. 114886, 2021. Disponível em: https://doi.org/10.1016/j.eswa.2021.114886. Acesso em: 31 de maio de 2025.

WANG, Q.; ZHAO, Y.; XU, Z.; ZHANG, W.; ZHANG, M. Integrating symmetry in attribute-based sentiment modeling with enhanced hesitant fuzzy scoring for personalized online product recommendations. Symmetry, v. 16, art. 1652, 2024. Disponível em: https://doi.org/10.3390/sym16121652. Acesso em: 2 de junho de 2025.

XU, C.; KECHADI, M.-T. An enhanced fake news detection system with fuzzy deep learning. IEEE Access, 2024. Disponível em: https://doi.org/10.1109/ACCESS.2024.3418340. Acesso em: 26 de maio de 2025.

XU, C.; KECHADI, M.-T. Fuzzy deep hybrid network for fake news detection. In: Proceedings of the 12th International Symposium on Information and Communication Technology (SOICT 2023). ACM, 2023. Disponível em: https://dl.acm.org/doi/fullHtml/10.1145/3628797.3628971. Acesso em: 30 abril de 2025.

ZHANG, Y.; LI, X.; WANG, Y. The longer the better? The impact of online review length on tourist decision-making. Tourism Economics, v. 29, n. 3, p. 789–805, 2023.