Combinar análise de sentimento e conteúdo para explicar emoções subtis em avaliações online de clientes de ginásios
DOI:
https://doi.org/10.47197/retos.v78.118658Palavras-chave:
Emoções, condicionamento físico, aprendizagem de máquina, avaliações online, classificação por estrelasResumo
Introdução: As emoções geradas pelos serviços influenciam a satisfação e a experiência do cliente. As avaliações online permitem identificar estes padrões emocionais através de técnicas de análise de texto.
Objectivo: Examinar como a relevância temática e o sentimento presentes nas avaliações digitais predizem emoções específicas dos utilizadores de ginásios, propondo uma abordagem baseada em “temas polarizados”.
Metodologia: Foram recolhidas 3.250 avaliações de 38 academias espanholas. O estudo aplicou uma abordagem de métodos mistos que combinava a modelação temática utilizando LDA, análise de sentimentos com TextBlob e algoritmos de aprendizagem automática (XGBoost), integrados com técnicas de explicabilidade SHAP. A interação entre temas e polaridade do sentimento foi utilizada para prever dez emoções específicas, como a alegria, a raiva, a tristeza e a confiança.
Resultados: A cordialidade da equipa, o custo-benefício e a higiene surgiram como temas particularmente preditivos. As avaliações positivas relacionadas com a equipa aumentaram emoções como a alegria e a confiança, enquanto os comentários sobre as ausências relacionadas com a COVID-19 foram associados a níveis mais elevados de raiva e tristeza. A abordagem dos “temas polarizados” gerou uma classificação emocional robusta, alcançando pontuações F1 acima de 0,84 para a maioria das emoções.
Discussão: Os resultados mostram que a combinação de modelação temática, análise de sentimentos e técnicas explicáveis permite a identificação precisa de quais os aspetos do serviço que desencadeiam emoções específicas, fornecendo uma estrutura interpretativa útil para os gestores do setor do fitness.
Conclusões: O método proposto constitui uma abordagem escalável e transparente para prever emoções discretas em avaliações. A sua aplicação pode melhorar o design de serviços e o alinhamento emocional em organizações orientadas para o bem-estar.
Referências
Acheampong, F. A., Wenyu, C., & Nunoo‐Mensah, H. (2020). Text‐based emotion detection: Advances, challenges, and opportunities. Engineering Reports, 2(7), e12189 https://doi.org/10.1002/eng2.12189
Aloufi, S. & Saddik, A.E. (2018). Sentiment Identification in Football-Specific Tweets. IEEE Access, 6, 78609-78621. https://doi.org/10.1109/ACCESS.2018.2885117
Alzate, M., Arce-Urriza, M., & Cebollada, J. (2022). Mining the text of online consumer reviews to ana-lyze brand image and brand positioning. Journal of Retailing and Consumer Services, 67, 102989. https://doi.org/10.1016/j.jretconser.2022.102989
Billings, A. C., Butterworth, M. L., & Turman, P. D. (2017). Communication and sport: Surveying the field. Sage publications.
Chaklader, R., & Parkinson, M. B. (2017). Data-driven sizing specification utilizing consumer text re-views. Journal of Mechanical Design, 139(11), 111406. https://doi.org/10.1115/1.4037476
Chen, X., Sun, S., Zhang, Z., Ma, Z., Wu, X., Li, H., ... & Zhang, K. (2022). Consumer shopping emotion and interest database: a unique database with a multimodal emotion recognition method for retail service robots to infer consumer shopping intentions better than humans. Journal of Electronic Imaging, 31(6), 061807-061807. https://doi.org/10.1117/1.JEI.31.6.061807
Chiu, M. C., & Lin, K. Z. (2018). Utilizing text mining and Kansei Engineering to support data-driven de-sign automation at conceptual design stage. Advanced Engineering Informatics, 38, 826-839. https://doi.org/10.1016/j.aei.2018.11.002
Fan, W., Liu, Y., Li, H., Tuunainen, V. K., & Lin, Y. (2022). Quantifying the effects of online review content structures on hotel review helpfulness. Internet Research, 32(7), 202-227. https://doi.org/10.1108/INTR-11-2019-0452
Fernandes, S., Panda, R., Venkatesh, V. G., Swar, B. N., & Shi, Y. (2022). Measuring the impact of online reviews on consumer purchase decisions–A scale development study. Journal of Retailing and Consumer Services, 68, 103066. https://doi.org/10.1016/j.jretconser.2022.103066
Guo, F., Cao, Y., Ding, Y., Liu, W., & Zhang, X. (2014). A multimodal measurement method of users’ emo-tional experiences shopping online. Human Factors and Ergonomics in Manufacturing & Ser-vice Industries, 25(5), 585-598. https://doi.org/10.1002/hfm.20577
Guo, J., Wang, X., & Wu, Y. (2020). Positive emotion bias: Role of emotional content from online cus-tomer reviews in purchase decisions. Journal of Retailing and Consumer Services, 52, 101891. https://doi.org/10.1016/j.jretconser.2019.101891
Hastie, T., Tibshirani, R., Friedman, J. H. (2009). "10. Boosting and Additive Trees". The Elements of Sta-tistical Learning (2nd ed.). New York: Springer. pp. 337–384. ISBN 978-0-387-84857-0. https://doi.org/10.1007/978-0-387-84858-7_10
Janssens, O., Verstockt, S., Mannens, E., Van Hoecke, S., & Van de Walle, R. (2014). Influence of weak labels for emotion recognition of tweets. In Mining Intelligence and Knowledge Exploration: Second International Conference, MIKE 2014, Cork, Ireland, December 10-12, 2014. Proceed-ings (pp. 108-118). Springer International Publishing. https://doi.org/10.1007/978-3-319-13817-6_12
Jia, S. (2019). Toward a better fitness club: Evidence from exerciser online rating and review using la-tent Dirichlet allocation and support vector machine. International Journal of Market Re-search, 61(1), 64-76. https://doi.org/10.1177/1470785318770571
Jiao, Y., & Qu, Q. X. (2019). A proposal for Kansei knowledge extraction method based on natural lan-guage processing technology and online product reviews. Computers in Industry, 108, 1-11. https://doi.org/10.1016/j.compind.2019.02.011
Kim, J., & Gupta, P. (2012). Emotional expressions in online user reviews: How they influence consum-ers' product evaluations. Journal of Business Research, 65(7), 985-992. https://doi.org/10.1016/j.jbusres.2011.04.013
Kranzbühler, A. M., Zerres, A., Kleijnen, M. H., & Verlegh, P. W. (2020). Beyond valence: A meta-analysis of discrete emotions in firm-customer encounters. Journal of the Academy of Marketing Sci-ence, 48, 478-498. https://doi.org/10.1007/s11747-019-00707-0
Kumar, P., Malik, S., & Raman, B. (2022). Hybrid fusion based interpretable multimodal emotion recog-nition with insufficient labelled data. arXiv preprint arXiv:2208.11450. https://doi.org/10.48550/arXiv.2208.11450
Kwak, D. , Kim, Y. & Hirt, E. (2011). Exploring the role of emotions on sport consumers' behavioral and cognitive responses to marketing stimuli. European Sport management quarterly, 11(3), 225-250. https://doi.org/10.1080/16184742.2011.577792
Lelieveld, G. J., & Hendriks, H. (2021). The interpersonal effects of distinct emotions in online reviews. Cognition and Emotion, 35(7), 1257-1280. https://doi.org/10.1080/02699931.2021.1947199
Li, L., Goh, T. T., & Jin, D. (2020). How textual quality of online reviews affect classification perfor-mance: a case of deep learning sentiment analysis. Neural Computing and Applications, 32, 4387-4415. https://doi.org/10.1007/s00521-018-3865-7
Li, X., Wu, C., & Mai, F. (2019). The effect of online reviews on product sales: A joint sentiment-topic analysis. Information & Management, 56(2), 172-184. https://doi.org/10.1016/j.im.2018.04.007
Lin, P. T., Vu, T. T., Nguyen, V. P., & Wu, Q. (2022). Self-determination theory and accountant employ-ees’ psychological wellbeing: The roles of positive affectivity and psychological safe-ty. Frontiers in Psychology, 13, 870771. https://doi.org/10.3389/fpsyg.2022.870771
Liu, H., Cui, T., & He, M. (2021). Product optimization design based on online review and orthogonal experiment under the background of big data. Proceedings of the Institution of Mechanical En-gineers, Part E: Journal of Process Mechanical Engineering, 235(1), 52-65. https://doi.org/10.1177/0954408920943690
Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, 30.
Lundberg, S. M., Nair, B., Vavilala, M. S., Horibe, M., Eisses, M. J., Adams, T., ... & Lee, S. I. (2018). Explaina-ble machine-learning predictions for the prevention of hypoxaemia during surgery. Nature bi-omedical engineering, 2(10), 749-760. https://doi.org/10.1038/s41551-018-0304-0
Mao, Y., Dai, C., Zhang, Y., Shen, S., Ma, H., & Chen, R. (2019). Impact of Review Emotions on Sales: The Moderating Role of Product Type. Review of Integrative Business and Economics Research, 8(4), 97.
Mathayomchan, B., & Taecharungroj, V. (2020). “How was your meal?” Examining customer experi-ence using Google maps reviews. International Journal of Hospitality Management, 90, 102641. https://doi.org/10.1016/j.ijhm.2020.102641
Murdoch, W. J., Singh, C., Kumbier, K., Abbasi-Asl, R., & Yu, B. (2019). Interpretable machine learning: definitions, methods, and applications. arXiv preprint arXiv:1901.04592. https://doi.org/10.48550/arXiv.1901.04592
Nusairat, N. & Hammouri, Q., Al-Ghadir, H., Ahmad, A. & Abuhashesh, M. (2020). Fitness centers ambi-ence-customer behavioral intentions relationship: The mediating role of customer emotional states. International Journal of Business and Management, 15(9), 93. https://doi.org/10.5539/ijbm.v15n9p93
Ozyurt, B., & Akcayol, M. A. (2021). A new topic modeling based approach for aspect extraction in as-pect based sentiment analysis: SS-LDA. Expert Systems with Applications, 168, 114231. https://doi.org/10.1016/j.eswa.2020.114231
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., ... & Duchesnay, É. (2011). Scikit-learn: Machine learning in Python. the Journal of machine Learning research, 12, 2825-2830. http://jmlr.org/papers/v12/pedregosa11a.html
Plutchik, R., & Conte, H. (1997). Circumplex models of personality and emotions. American Psychologi-cal Association.
Rathore, A. K., & Ilavarasan, P. V. (2020). Pre-and post-launch emotions in new product development: Insights from twitter analytics of three products. International Journal of Information Man-agement, 50, 111-127. https://doi.org/10.1016/j.ijinfomgt.2019.05.015
Rocklage, M., Rucker, D., & Nordgren, L. (2021). Mass-scale emotionality reveals human behaviour and marketplace success. Nature human behaviour, 5(10), 1323-1329. https://doi.org/10.1038/s41562-021-01098-5
Russel, J.A.: A Circumplex Model of Affect, Journal of Personality and Social Psychology, 39(6), 1161-1178, 1980. https://doi.org/10.1037/h0077714
Samuel, J., Rozzi, G., & Palle, R. (2022). The dark side of sentiment analysis: An exploratory review using lexicons, dictionaries, and a statistical monkey and chimp. Dictionaries, and a Statistical Monkey and Chimp (January 6, 2022). http://dx.doi.org/10.2139/ssrn.4000087
Seyeditabari, A., Tabari, N., Gholizadeh, S., & Zadrozny, W. (2019). Emotion detection in text: focusing on latent representation. arXiv preprint arXiv:1907.09369. https://doi.org/10.48550/arXiv.1907.09369
Sievert, C., & Shirley, K. (2014, June). LDAvis: A method for visualizing and interpreting topics. In Proceedings of the workshop on interactive language learning, visualization, and interfac-es (pp. 63-70). https://aclanthology.org/W14-3110.pdf
Shah, A. M., Abbasi, A. Z., & Yan, X. (2023). Do online peer reviews stimulate diners’ continued log-in behavior: Investigating the role of emotions in the O2O meal delivery apps context. Journal of Retailing and Consumer Services, 72, 103234. https://doi.org/10.1016/j.jretconser.2022.103234
Shaver, P., Schwartz, J., Kirson, D., & O'Connor, C. (1987). Emotion knowledge: Further exploration of a prototype approach. Journal of Personality and Social Psychology, 52(6), 1061–1086. https://doi.org/10.1037/0022-3514.52.6.1061
So, C. (2020). Understanding the prediction mechanism of sentiments by XAI visualization. In Proceedings of the 4th international conference on natural language processing and infor-mation retrieval (75-80). https://doi.org/10.1145/3443279.3443284
Štajner, S. (2021, september). Exploring Reliability of Gold Labels for Emotion Detection in Twitter. [Paper] International Conference on Recent Advances in Natural Language Processing, 1350-1359 Online. https://doi.org/10.26615/978-954-452-072-4_151
Tafreshi, S., & Diab, M. (2018). Emotion Detection and Classification in a Multigenre Corpus with Joint Multi-Task Deep Learning. In Proceedings of the 27th International Conference on Computa-tional Linguistics, pages 2905–2913, Santa Fe, New Mexico, USA. Association for Computational Linguistics. https://aclanthology.org/C18-1246
Wang, W., Chen, L., Thirunarayan, K., & Sheth, A. P. (2012, 3-5 september). Harnessing twitter" big da-ta" for automatic emotion identification. [Proceeding] 2012 International Conference on Priva-cy, Security, Risk and Trust and 2012 International Conferenece on Social Computing (pp. 587-592), Amsterdam Netherlands. https://doi.org/10.1109/PASSAT-SocialCom30175.2012
Wang, Y., Lu, X., & Tan, Y. (2018). Impact of product attributes on customer satisfaction: An analysis of online reviews for washing machines. Electronic commerce research and applications, 29, 1-11. https://doi.org/10.1016/j.elerap.2018.03.003
Wani, M. A., Bours, P., Agarwal, N., & Jabin, S. (2019). Emotion-based mining for gender prediction in online social networks. In Proceedings of the ACM, International Conference on Machine Learn-ing and Data Science.
Xiang, M., Zhong, D., Han, M., & Lv, K. (2023, July). A Study on Online Health Community Users’ Infor-mation Demands Based on the BERT-LDA Model. Healthcare, 11(15), 2142. https://doi.org/10.3390/healthcare11152142
Xu, X. (2019). Examining the relevance of online customer textual reviews on hotels’ product and ser-vice attributes. Journal of Hospitality & Tourism Research, 43(1), 141-163. https://doi.org/10.1177/1096348018764573
Yi, Q., Khan, J., Su, Y., Tong, J., & Zhao, S. (2023). Impulse buying tendency in live-stream commerce: The role of viewing frequency and anticipated emotions influencing scarcity-induced purchase decision. Journal of Retailing and Consumer Services, 75, 103534.https://doi.org/10.1016/j.jretconser.2023.103534.
Zhai, Z., Chen, H., Feng, F., Li, R., & Wang, X. (2022, December). COM-MRC: A COntext-masked machine reading comprehension framework for aspect sentiment triplet extraction. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (pp. 3230-3241). https://doi.org/10.18653/v1/2022.emnlp-main.212
Zhang, W., Li, X., Deng, Y., Bing, L., & Lam, W. (2022). A survey on aspect-based sentiment analysis: Tasks, methods, and challenges. IEEE Transactions on Knowledge and Data Engineering, 35(11), 1101911038.https://doi.org/10.1109/TKDE.2022.3230975
Downloads
Publicado
Edição
Secção
Licença
Direitos de Autor (c) 2026 Enrique Alcántara Alcover, Antonio Hernández Martín, Ricardo Cuevas Campos, Daniel Duclos Bastías
Este trabalho encontra-se publicado com a Licença Internacional Creative Commons Atribuição-NãoComercial-SemDerivações 4.0.
Autores que publicam nesta revista concordam com os seguintes termos:
- Autores mantém os direitos autorais e assegurar a revista o direito de ser a primeira publicação da obra como licenciado sob a Licença Creative Commons Attribution que permite que outros para compartilhar o trabalho com o crédito de autoria do trabalho e publicação inicial nesta revista.
- Os autores podem estabelecer acordos adicionais separados para a distribuição não-exclusiva da versão do trabalho publicado na revista (por exemplo, a um repositório institucional, ou publicá-lo em um livro), com reconhecimento de autoria e publicação inicial nesta revista.
- É permitido e os autores são incentivados a divulgar o seu trabalho por via electrónica (por exemplo, em repositórios institucionais ou no seu próprio site), antes e durante o processo de envio, pois pode gerar alterações produtivas, bem como a uma intimação mais Cedo e mais do trabalho publicado (Veja O Efeito do Acesso Livre) (em Inglês).
Esta revista é a "política de acesso aberto" de Boai (1), apoiando os direitos dos usuários de "ler, baixar, copiar, distribuir, imprimir, pesquisar, ou link para os textos completos dos artigos". (1) http://legacy.earlham.edu/~peters/fos/boaifaq.htm#openaccess
