An Evaluation of Meta-Features for Automated Detection of Persuasion in Texts of Political Memes
Resumo
This work proposes a feature-engineering-based approach for detecting persuasive strategies in political memes. Four groups of meta-features were designed: (i) rhetorical, (ii) sentiment and hate speech, (iii) structural, and (iv) contextual features. Experiments used the SemEval-2024 Task 4 dataset with 7,000 training and 1,000 testing instances. Random Forest and Logistic Regression were evaluated with and without SMOTE to handle class imbalance present on the training dataset. The best result, with a macro-F1 value of 0.698, was achieved by combining rhetorical and structural features. The proposed approach offers a computationally efficient and interpretable alternative to neural-based models.
Palavras-chave:
Persuasion Classification, Meta-Features, Aristotelian Rhetoric, Text Classification, Memes
Referências
Anghelina, I., But, , G., and Enache, A. SuteAlbastre at SemEval-2024 task 4: Predicting propaganda techniques in multilingual memes using joint text and vision transformers. In Proc. of the 18th Int'l. Wksh on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 443 -449, 2024.
Braca, A. and Dondio, P. Developing persuasive systems for marketing: the interplay of persuasion techniques, customer traits and persuasive message design. Ital. J. Mark vol. 2023, pp. 369 -412, 2023.
Carvalho, J. and Plastino, A. On the evaluation and combination of state-of-the-art features in twitter sentiment analysis. Artif. Intell. Rev. 54 (3): 1887 -1936, 2021.
Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. Smote: synthetic minority over-sampling technique. J. Artif. Int. Res. 16 (1): 321 -357, 2002.
Cruz, A. F., Rocha, G., and Cardoso, H. L. On sentence representations for propaganda detection: From handcrafted features to word embeddings. In Proc. of the 2nd Wkshp on Natural Language Processing for Internet Freedom: Censorship, Disinformation, and Propaganda. ACL, Hong Kong, China, pp. 107 -112, 2019.
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proc. of the 2019 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). ACL, Minneapolis, Minnesota, pp. 4171 -4186, 2019.
Dimitrov, D., Alam, F., Hasanain, M., Hasnat, A., Silvestri, F., Nakov, P., and Da San Martino, G. SemEval-2024 task 4: Multilingual detection of persuasion techniques in memes. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 2009 -2026, 2024.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., and Houlsby, N. An image is worth 16x16 words: Transformers for image recognition at scale. ArXiv vol. abs/2010.11929, 2020.
Ferreira, T. C. and Paraboni, I. Geração de linguagem natural. In Processamento de Linguagem Natural: Conceitos, Técnicas e Aplicações em Português, 3 ed., H. M. Caseli and M. G. V. Nunes (Eds.). BPLN, Book chapter 16, 2024.
Guerini, M., Özbal, G., and Strapparava, C. Echoes of persuasion: The eect of euphony in persuasive communication. In Proc of the 2015 HLT-NAACL Conf. ACL, Denver, Colorado, pp. 1483 -1493, 2015.
Halversen, A. and Weeks, B. E. Memeing politics: Understanding political meme creators, audiences, and consequences on social media. Social Media + Society 9 (4): 20563051231205588, 2023.
Hu, M. and Liu, B. Mining and summarizing customer reviews. In Proc. of the 10th ACM SIGKDD Int'l. Conf. on Knowledge Discovery and Data Mining. KDD '04. ACM, New York, NY, USA, pp. 168 -177, 2004.
Li, S., Wang, Y., Yang, L., Zhang, S., and Lin, H. LMEME at SemEval-2024 task 4: Teacher student fusion - integrating CLIP with LLMs for enhanced persuasion detection. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 628 -633, 2024.
Mohamad, H. A. Analysis of rhetorical appeals to logos, ethos and pathos in enl and esl research abstracts. Malaysian Journal of Social Sciences and Humanities 7 (3), 2022.
Navigli, R., Conia, S., and Ross, B. Biases in large language models: Origins, inventory, and discussion. J. Data and Information Quality 15 (2), 2023.
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., and Sutskever, I. Learning transferable visual models from natural language supervision. In Proc. of the 38th Int'l. Conf. on Machine Learning. Vol. 139. PMLR, pp. 8748 -8763, 2021.
Takahashi, H. Hidetsune at SemEval-2024 task 4: An application of machine learning to multilingual propagandistic memes identication using machine translation. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 370 -373, 2024.
Vargas, F., Carvalho, I., Pardo, T. A. S., and Benevenuto, F. Context-aware and expert data resources for brazilian portuguese hate speech detection. Natural Language Processing 31 (2): 435 -456, 2025.
Weiss, M. C. Sociedade sensoriada: a sociedade da transformação digital. Estudos Avançados 33 (95): 203 -214, 2019.
Yu, E., Wang, J., Qiao, X., Qi, J., Li, Z., Lin, H., Zong, L., and Xu, B. DUTIR938 at SemEval-2024 task 4: Semi-supervised learning and model ensemble for persuasion techniques detection in memes. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 642 -648, 2024
Braca, A. and Dondio, P. Developing persuasive systems for marketing: the interplay of persuasion techniques, customer traits and persuasive message design. Ital. J. Mark vol. 2023, pp. 369 -412, 2023.
Carvalho, J. and Plastino, A. On the evaluation and combination of state-of-the-art features in twitter sentiment analysis. Artif. Intell. Rev. 54 (3): 1887 -1936, 2021.
Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. Smote: synthetic minority over-sampling technique. J. Artif. Int. Res. 16 (1): 321 -357, 2002.
Cruz, A. F., Rocha, G., and Cardoso, H. L. On sentence representations for propaganda detection: From handcrafted features to word embeddings. In Proc. of the 2nd Wkshp on Natural Language Processing for Internet Freedom: Censorship, Disinformation, and Propaganda. ACL, Hong Kong, China, pp. 107 -112, 2019.
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proc. of the 2019 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). ACL, Minneapolis, Minnesota, pp. 4171 -4186, 2019.
Dimitrov, D., Alam, F., Hasanain, M., Hasnat, A., Silvestri, F., Nakov, P., and Da San Martino, G. SemEval-2024 task 4: Multilingual detection of persuasion techniques in memes. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 2009 -2026, 2024.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., and Houlsby, N. An image is worth 16x16 words: Transformers for image recognition at scale. ArXiv vol. abs/2010.11929, 2020.
Ferreira, T. C. and Paraboni, I. Geração de linguagem natural. In Processamento de Linguagem Natural: Conceitos, Técnicas e Aplicações em Português, 3 ed., H. M. Caseli and M. G. V. Nunes (Eds.). BPLN, Book chapter 16, 2024.
Guerini, M., Özbal, G., and Strapparava, C. Echoes of persuasion: The eect of euphony in persuasive communication. In Proc of the 2015 HLT-NAACL Conf. ACL, Denver, Colorado, pp. 1483 -1493, 2015.
Halversen, A. and Weeks, B. E. Memeing politics: Understanding political meme creators, audiences, and consequences on social media. Social Media + Society 9 (4): 20563051231205588, 2023.
Hu, M. and Liu, B. Mining and summarizing customer reviews. In Proc. of the 10th ACM SIGKDD Int'l. Conf. on Knowledge Discovery and Data Mining. KDD '04. ACM, New York, NY, USA, pp. 168 -177, 2004.
Li, S., Wang, Y., Yang, L., Zhang, S., and Lin, H. LMEME at SemEval-2024 task 4: Teacher student fusion - integrating CLIP with LLMs for enhanced persuasion detection. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 628 -633, 2024.
Mohamad, H. A. Analysis of rhetorical appeals to logos, ethos and pathos in enl and esl research abstracts. Malaysian Journal of Social Sciences and Humanities 7 (3), 2022.
Navigli, R., Conia, S., and Ross, B. Biases in large language models: Origins, inventory, and discussion. J. Data and Information Quality 15 (2), 2023.
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., and Sutskever, I. Learning transferable visual models from natural language supervision. In Proc. of the 38th Int'l. Conf. on Machine Learning. Vol. 139. PMLR, pp. 8748 -8763, 2021.
Takahashi, H. Hidetsune at SemEval-2024 task 4: An application of machine learning to multilingual propagandistic memes identication using machine translation. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 370 -373, 2024.
Vargas, F., Carvalho, I., Pardo, T. A. S., and Benevenuto, F. Context-aware and expert data resources for brazilian portuguese hate speech detection. Natural Language Processing 31 (2): 435 -456, 2025.
Weiss, M. C. Sociedade sensoriada: a sociedade da transformação digital. Estudos Avançados 33 (95): 203 -214, 2019.
Yu, E., Wang, J., Qiao, X., Qi, J., Li, Z., Lin, H., Zong, L., and Xu, B. DUTIR938 at SemEval-2024 task 4: Semi-supervised learning and model ensemble for persuasion techniques detection in memes. In Proc. of the 18th Int'l. Wkshp on Semantic Evaluation (SemEval-2024). ACL, Mexico City, Mexico, pp. 642 -648, 2024
Publicado
29/09/2025
Como Citar
AZEVEDO, Ana B. S. de; GONÇALVES, Eduardo C..
An Evaluation of Meta-Features for Automated Detection of Persuasion in Texts of Political Memes. In: SYMPOSIUM ON KNOWLEDGE DISCOVERY, MINING AND LEARNING (KDMILE), 13. , 2025, Fortaleza/CE.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 145-152.
ISSN 2763-8944.
DOI: https://doi.org/10.5753/kdmile.2025.247776.
