Identification of explicit and implicit aspects in culinary reviews in Portuguese: evaluating the potential of LLMs

• Luiz H. N. Silva IFSP http://orcid.org/0009-0007-2839-8259
• Eloize R. M. Seno IFSP https://orcid.org/0000-0002-1549-9794
• Rozane R. Rebechi UFRGS https://orcid.org/0000-0002-1878-7548
• Helena M. Caseli UFSCar https://orcid.org/0000-0003-3996-8599
• Fabiano M. Rocha Júnior IFSP
• Guilherme A. Faller UFRGS

DOI: https://doi.org/10.5753/stil.2024.245360

Abstract

Aspect identification is a fundamental step in Aspect-Based Sentiment Analysis (ABSA), which involves detecting the opinion target aspects in product or service reviews published on social media. Although there are many works developed for detecting aspects in English, there are few studies in this area for Portuguese, and LLMs have been little explored. Given this context, this research investigated the potential use of LLMs for aspect identification in culinary reviews in Portuguese.

Keywords: Aspect-based sentiment analysis, explicit and implicit aspects, LLMs

References

Almeida, T. S., Abonizio, H., Nogueira, R., and Pires, R. (2024). Sabiá-2: A new generation of portuguese large language models. ArXiv, abs/2403.09887.

Assi, F. M., Candido, G. B., dos Santos Silva, L. N., Silva, D. F., and Caseli, H. M. (2022). Ufscar’s team at ABSAPT 2022: using syntax, semantics and context for solving the tasks. In Montes-y-Gomez, M. and et al., editors, Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2022), volume 3202 of CEUR Workshop Proceedings. CEUR-WS.org.

Balage Filho, P. P. (2017). Aspect extraction in sentiment analysis for portuguese language. PhD thesis, Sao Carlos - SP.

Costa, R. and Pardo, T. (2020). Métodos baseados em léxico para extração de aspectos de opiniões em português. In Anais do IX Brazilian Workshop on Social Network Analysis and Mining, pages 61–72, Porto Alegre, RS, Brasil. SBC.

Lopes, E., Correa, U., and Freitas, L. (2021). Exploring BERT for aspect extraction in portuguese language. The International FLAIRS Conference Proceedings, 34.

Machado, M., Pardo, T., Ruiz, E., and Felippo, A. (2021). Learning rules for automatic identification of implicit aspects in portuguese. In Anais do XIII Simposio Brasileiro de Tecnologia da Informação e da Linguagem Humana, pages 82–91, Porto Alegre, RS, Brasil. SBC.

Machado, M. and Pardo, T. A. S. (2022). Evaluating methods for extraction of aspect terms in opinion texts in Portuguese - the challenges of implicit aspects. In Calzolari, N., Bechet, F., Blache, P., Choukri, K., Cieri, C., Declerck, T., Goggi, S., Isahara, H., Maegaard, B., Mariani, J., Mazo, H., Odijk, J., and Piperidis, S., editors, Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 3819–3828, Marseille, France. European Language Resources Association.

Machado, M. T. (2023). Methods for identifying aspects in opinion texts in Portuguese: the case of implicit aspects and their typological analysis. PhD thesis, Sao Carlos - SP.

Oliveira, A., Cecote, T., Silva, P., Gertrudes, J., Freitas, V., and Luz, E. (2023). How good is ChatGPT for detecting hate speech in portuguese? In Anais do XIV Simpósio Brasileiro de Tecnologia da Informação e da Linguagem Humana, pages 94–103, Porto Alegre, RS, Brasil. SBC.

Pereira, D. A. (2021). A survey of sentiment analysis in the portuguese language. Artificial Intelligence Review, 54(2):1087–1115.

Rebechi, R. R., Nunes, R. R., Munhoz, L. R., and Marcon, N. O. (2021). Restaurant reviews in Brazil and the USA: A feast of cultural differences and their impact on translation. Mutatis Mutandis. Revista Latinoamericana de Traduccion, 14:372–396.

Resplande, J., Garcia, E., Junior, A., Rodrigues, R., Silva, D., Maia, D., Da Silva, N., Filho, A., and Soares, A. (2022). Deep learning Brasil at ABSAPT 2022: Portuguese transformer ensemble approaches. In Montes-y-Gomez, M. and et al., editors, Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2022), volume 3202 of CEUR Workshop Proceedings. CEUR-WS.org.

Santos, W. and Paraboni, I. (2023). Predição de transtorno depressivo em redes sociais: BERT supervisionado ou ChatGPT zero-shot? In Anais do XIV Simpósio Brasileiro de Tecnologia da Informação e da Linguagem Humana, pages 11–21, Porto Alegre, RS, Brasil. SBC.

Schouten, K. and Frasincar, F. (2016). Survey on aspect-level sentiment analysis. IEEE Transactions on Knowledge and Data Engineering, 28(3):813–830.

Seno, E., Silva, L., Anno, F., Rocha, F., and Caseli, H. (2024). Aspect-based sentiment analysis in comments on political debates in Portuguese: evaluating the potential of ChatGPT. In Gamallo, P., Claro, D., Teixeira, A., Real, L., Garcia, M., Oliveira, H. G., and Amaro, R., editors, Computational Processing of the Portuguese Language: 16th Conference, PROPOR 2024, pages 312–320, Santiago de Compostela, Galicia/Spain. Association for Computational Lingustics.

Soni, P. K. and Rambola, R. (2022). A survey on implicit aspect detection for sentiment analysis: Terminology, issues, and scope. IEEE Access, 10:63932–63957.

Vargas, F. A. and Pardo, T. A. S. (2018). Aspect clustering methods for sentiment analysis. In Computational Processing of the Portuguese Language: 13th International Conference, PROPOR 2018, Canela, Brazil, September 24–26, 2018, Proceedings, page 365–374, Berlin, Heidelberg. Springer-Verlag.

Vargas, F. A. and Pardo, T. A. S. (2020). Linguistic rules for fine-grained opinion extraction. proceedings of the 14th International AAAI Conference on Web and Social Media, 2020. Zhang, L., Wang, S., and Liu, B. (2018). Deep learning for sentiment analysis : A survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 8

Zhang, L., Wang, S., and Liu, B. (2018). Deep learning for sentiment analysis : A survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 8.