Sentimento vs. Nota: Detectando Discrepâncias em Avaliações de Aplicativos
Resumo
As avaliações online exercem forte influência no comportamento dos consumidores, mas costumam apresentar divergências entre a nota numérica e o sentimento expresso no texto. Este artigo apresenta uma análise quantitativa dessas discrepâncias utilizando modelos de aprendizado de máquina aplicados ao UTL-Corpus, um córpus de avaliações da Google Play Store. Dentre os modelos testados para classificação de polaridade, o BERTimbau obteve melhor desempenho (acurácia de 91% e F1-score de 90%) e identificou 1.772 avaliações discrepantes (17,7% do total), em sua maioria positivas. O estudo reforça a importância de analisar o conteúdo textual além das notas para aprimorar a compreensão de feedbacks e a eficácia dos sistemas de recomendação.Referências
Ahn, Y. and Lee, J. (2024). The impact of online reviews on consumers’ purchase intentions: Examining the social influence of online reviews, group similarity, and self-construal. Journal of Theoretical and Applied Electronic Commerce Research, 19:1060–1078.
Almansour, A., Al-Otaibi, R., and Alharbi, H. (2022). Text-rating review discrepancy (trrd): an integrative review and implications for research. Future Business Journal, 8.
Brum, H. B. and das Graças Volpe Nunes, M. (2017). Building a sentiment corpus of tweets in brazilian portuguese. arXiv preprint arXiv:1712.08917.
Chang, H.-L., Liu, Y.-L., Keng, C.-J., and Jiang, H.-L. (2024). Examining discrepancies between online product ratings and sentiments expressed in review contents. Management Analytics and Social Insights, 1(1):129–144.
Chen, T., Samaranayake, P., Cen, X., Qi, M., and Lan, Y.-C. (2022). The impact of online reviews on consumers’ purchasing decisions: Evidence from an eye-tracking study. Frontiers in Psychology, 13.
Constantinides, E. and Holleschovsky, N. (2016). Impact of online product reviews on purchasing decisions. pages 271–278.
Fazzolari, M., Cozza, V., Petrocchi, M., and Spognardi, A. (2017). A study on text-score disagreement in online reviews. Cognitive Computation.
Sadiq, S., Umer, M., Ullah, D. S., Mirjalili, S., Rupapara, V., and NAPPI, M. (2021). Discrepancy detection between actual user reviews and numeric ratings of google app store using deep learning. Expert Systems with Applications, 181:115111.
Shan, G., Zhou, L., and Zhang, D. (2021). From conflicts and confusion to doubts: Examining review inconsistency for fake review detection. Decision Support Systems, 144:113513.
Sousa, R. F. d. (2023). Classificação da utilidade de opiniões em português brasileiro. PhD thesis, Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, São Carlos.
Souza, F., Nogueira, R., and Lotufo, R. (2020). Bertimbau: Pretrained bert models for brazilian portuguese. In Cerri, R. and Prati, R. C., editors, Intelligent Systems, pages 403–417, Cham. Springer International Publishing.
Xu, Z., Zeng, H., and Ai, Q. (2021). Understanding the effectiveness of reviews in e-commerce top-n recommendation. ICTIR ’21: Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval.
Almansour, A., Al-Otaibi, R., and Alharbi, H. (2022). Text-rating review discrepancy (trrd): an integrative review and implications for research. Future Business Journal, 8.
Brum, H. B. and das Graças Volpe Nunes, M. (2017). Building a sentiment corpus of tweets in brazilian portuguese. arXiv preprint arXiv:1712.08917.
Chang, H.-L., Liu, Y.-L., Keng, C.-J., and Jiang, H.-L. (2024). Examining discrepancies between online product ratings and sentiments expressed in review contents. Management Analytics and Social Insights, 1(1):129–144.
Chen, T., Samaranayake, P., Cen, X., Qi, M., and Lan, Y.-C. (2022). The impact of online reviews on consumers’ purchasing decisions: Evidence from an eye-tracking study. Frontiers in Psychology, 13.
Constantinides, E. and Holleschovsky, N. (2016). Impact of online product reviews on purchasing decisions. pages 271–278.
Fazzolari, M., Cozza, V., Petrocchi, M., and Spognardi, A. (2017). A study on text-score disagreement in online reviews. Cognitive Computation.
Sadiq, S., Umer, M., Ullah, D. S., Mirjalili, S., Rupapara, V., and NAPPI, M. (2021). Discrepancy detection between actual user reviews and numeric ratings of google app store using deep learning. Expert Systems with Applications, 181:115111.
Shan, G., Zhou, L., and Zhang, D. (2021). From conflicts and confusion to doubts: Examining review inconsistency for fake review detection. Decision Support Systems, 144:113513.
Sousa, R. F. d. (2023). Classificação da utilidade de opiniões em português brasileiro. PhD thesis, Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, São Carlos.
Souza, F., Nogueira, R., and Lotufo, R. (2020). Bertimbau: Pretrained bert models for brazilian portuguese. In Cerri, R. and Prati, R. C., editors, Intelligent Systems, pages 403–417, Cham. Springer International Publishing.
Xu, Z., Zeng, H., and Ai, Q. (2021). Understanding the effectiveness of reviews in e-commerce top-n recommendation. ICTIR ’21: Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval.
Publicado
20/07/2025
Como Citar
MOURA FILHO, Marcos Paulo Soares; SOUSA, Rogério F. de.
Sentimento vs. Nota: Detectando Discrepâncias em Avaliações de Aplicativos. In: BRAZILIAN WORKSHOP ON SOCIAL NETWORK ANALYSIS AND MINING (BRASNAM), 14. , 2025, Maceió/AL.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 297-303.
ISSN 2595-6094.
DOI: https://doi.org/10.5753/brasnam.2025.9308.
