{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T05:40:42Z","timestamp":1765172442353,"version":"3.46.0"},"reference-count":68,"publisher":"Emerald","issue":"6","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,12,9]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>In this paper, we address the need to study automatic propaganda detection to establish a course of action when faced with such a complex task. Although many isolated tasks have been proposed, a roadmap on how to best approach a new task from the perspective of text formality or the leverage of existing resources has not been explored yet.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Design\/methodology\/approach<\/jats:title>\n                    <jats:p>We present a comprehensive study using several datasets on textual propaganda and different techniques to tackle it. We explore diverse collections with varied characteristics and analyze methodologies, from classic machine learning algorithms, to multi-task learning to utilize the available data in such models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Findings<\/jats:title>\n                    <jats:p>Our results show that transformer-based approaches are the best option with high-quality collections, and emotionally enriched inputs improve the results for Twitter content. Additionally, MTL achieves the best results in two of the five scenarios we analyzed. Notably, in one of the scenarios, the model achieves an F1 score of 0.78, significantly surpassing the transformer baseline model\u2019s F1 score of 0.68.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Research limitations\/implications<\/jats:title>\n                    <jats:p>After finding a positive impact when leveraging propaganda\u2019s emotional content, we propose further research into exploiting other complex dimensions, such as moral issues or logical reasoning.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Originality\/value<\/jats:title>\n                    <jats:p>Based on our findings, we provide a roadmap for tackling propaganda-related tasks, depending on the types of training data available and the task to solve. This includes the application of MTL, which has yet to be fully exploited in propaganda detection.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1108\/intr-05-2024-0785","type":"journal-article","created":{"date-parts":[[2024,10,28]],"date-time":"2024-10-28T04:09:52Z","timestamp":1730088592000},"page":"2122-2145","source":"Crossref","is-referenced-by-count":0,"title":["Together we can do it! A roadmap to\u00a0effectively tackle propaganda-related tasks"],"prefix":"10.1108","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-6964-5956","authenticated-orcid":true,"given":"Raquel","family":"Rodr\u00edguez-Garc\u00eda","sequence":"first","affiliation":[{"name":"UNED NLP and IR Group, , ,","place":["Madrid, Spain"]}]},{"given":"Roberto","family":"Centeno","sequence":"additional","affiliation":[{"name":"UNED NLP and IR Group, , ,","place":["Madrid, Spain"]}]},{"given":"\u00c1lvaro","family":"Rodrigo","sequence":"additional","affiliation":[{"name":"UNED NLP and IR Group, , ,","place":["Madrid, Spain"]}]}],"member":"140","published-online":{"date-parts":[[2024,10,29]]},"reference":[{"key":"2025120800370056400_ref001","first-page":"301","article-title":"Detecting propaganda techniques in English news articles using pre-trained transformers","author":"Abdullah","year":"2022"},{"issue":"2","key":"2025120800370056400_ref002","doi-asserted-by":"publisher","first-page":"956","DOI":"10.11591\/ijai.v12.i2.pp956-965","article-title":"Combating propaganda texts using transfer learning","volume":"12","author":"Abdullah","year":"2023","journal-title":"International Journal of Artificial Intelligence"},{"key":"2025120800370056400_ref003","doi-asserted-by":"publisher","first-page":"2623","DOI":"10.1145\/3292500.3330701","article-title":"Optuna: a next-generation hyperparameter optimization framework","author":"Akiba","year":"2019"},{"key":"2025120800370056400_ref004","first-page":"108","article-title":"Overview of the WANLP 2022","author":"Alam","year":"2022"},{"issue":"3","key":"2025120800370056400_ref005","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3233\/kes-230192","article-title":"ArMT-TNN: enhancing natural language understanding performance through hard parameter multitask learning in Arabic","volume":"28","author":"Alkhathlan","year":"2024","journal-title":"International Journal of Knowledge-Based and Intelligent Engineering Systems"},{"key":"2025120800370056400_ref006","doi-asserted-by":"publisher","first-page":"5809","DOI":"10.18653\/v1\/2022.acl-long.399","article-title":"Evaluating extreme hierarchical multi-label classification","author":"Amigo","year":"2022"},{"key":"2025120800370056400_ref007","first-page":"1","article-title":"Noisy text data: foible of popular transformer based NLP models","author":"Bagla","year":"2024","journal-title":"Proceedings of the Third International Conference on AI-ML Systems"},{"key":"2025120800370056400_ref008","doi-asserted-by":"publisher","first-page":"1395","DOI":"10.18653\/v1\/2023.semeval-1.193","article-title":"Kb at SemEval-2023 task 3: on multitask hierarchical BERT base neural network for multi-label persuasion techniques detection","author":"Baraniak","year":"2023"},{"issue":"5","key":"2025120800370056400_ref009","doi-asserted-by":"publisher","first-page":"1849","DOI":"10.1016\/j.ipm.2019.03.005","article-title":"Proppy: organizing the news based on their propagandistic content","volume":"56","author":"Barr\u00f3n-Cede\u00f1o","year":"2019","journal-title":"Information Processing and Management"},{"issue":"1","key":"2025120800370056400_ref010","doi-asserted-by":"publisher","DOI":"10.1016\/j.ipm.2022.103118","article-title":"Detecting racial stereotypes: an Italian social media corpus where psychology meets NLP","volume":"60","author":"Bosco","year":"2023","journal-title":"Information Processing and Management"},{"issue":"2","key":"2025120800370056400_ref011","doi-asserted-by":"publisher","first-page":"388","DOI":"10.2307\/3647684","article-title":"Striking a responsive chord: how political ads motivate and persuade voters by appealing to emotions","volume":"49","author":"Brader","year":"2005","journal-title":"American Journal of Political Science"},{"issue":"1","key":"2025120800370056400_ref012","doi-asserted-by":"publisher","DOI":"10.1038\/s42005-020-0340-4","article-title":"The role of bot squads in the political propaganda on Twitter","volume":"3","author":"Caldarelli","year":"2020","journal-title":"Communications Physics"},{"key":"2025120800370056400_ref013","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1016\/b978-1-55860-307-3.50012-5","article-title":"Multitask learning: a knowledge-based source of inductive bias","author":"Caruana","year":"1993"},{"key":"2025120800370056400_ref014","first-page":"1","article-title":"PropaLTL at DIPROMATS: incorporating contextual features with BERT\u2019s auxiliary input for propaganda detection on tweets","author":"Casavantes","year":"2023"},{"key":"2025120800370056400_ref015","first-page":"1","article-title":"Dataset of propaganda techniques of the state-sponsored information operation of the people\u2019s Republic of China","author":"Chang","year":"2021","journal-title":"ArXiv preprint"},{"issue":"4","key":"2025120800370056400_ref016","doi-asserted-by":"publisher","first-page":"175","DOI":"10.3390\/bdcc7040175","article-title":"Empowering propaganda detection in resource-restraint languages: a transformer-based framework for classifying Hindi news articles","volume":"7","author":"Chaudhari","year":"2023","journal-title":"Big Data and Cognitive Computing"},{"key":"2025120800370056400_ref017","unstructured":"Chollet, F.\n           (2015), \u201cKeras\u201d, available at:\u00a0https:\/\/keras.io (accessed\u00a01 September 2024)."},{"key":"2025120800370056400_ref018","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/icassp49357.2023.10096771","article-title":"Multimodal propaganda detection via anti-persuasion prompt enhanced contrastive learning","author":"Cui","year":"2023"},{"key":"2025120800370056400_ref019","first-page":"162","article-title":"Findings of the NLP4IF-2019 shared task on fine-grained propaganda detection","author":"Da San Martino","year":"2019"},{"key":"2025120800370056400_ref020","first-page":"5636","article-title":"Fine-grained analysis of propaganda in news article","author":"Da San Martino","year":"2019"},{"key":"2025120800370056400_ref021","doi-asserted-by":"publisher","first-page":"1377","DOI":"10.18653\/v1\/2020.semeval-1.186","article-title":"SemEval-2020 task 11: detection of propaganda techniques in news articles","author":"Da San Martino","year":"2020"},{"key":"2025120800370056400_ref022","first-page":"287","article-title":"Prta: a system to Support the analysis of propaganda techniques in the news","author":"Da San Martino","year":"2020"},{"key":"2025120800370056400_ref023","first-page":"4826","article-title":"A survey on computational propaganda detection","author":"Da San Martino","year":"2021"},{"key":"2025120800370056400_ref024","first-page":"70","article-title":"SemEval-2021 task 6: detection of persuasion techniques in texts and images","author":"Dimitrov","year":"2021"},{"volume-title":"Propaganda: The Formation of Men\u2019s Attitudes","year":"1973","author":"Ellul","key":"2025120800370056400_ref025"},{"key":"2025120800370056400_ref026","doi-asserted-by":"publisher","first-page":"2650","DOI":"10.18653\/v1\/2023.findings-emnlp.175","article-title":"In what languages are generative language models the most formal? Analyzing formality distribution across languages","author":"Ersoy","year":"2023"},{"key":"2025120800370056400_ref027","first-page":"1","article-title":"Social media influencers and the 2020 US election: paying \u2018regular people\u2019 for digital campaign communication","author":"Goodwin","year":"2020","journal-title":"Center for Media Engagement"},{"key":"2025120800370056400_ref028","first-page":"845","article-title":"SemEval-2019 task 7: RumourEval 2019: determining rumour veracity and support for rumours","author":"Gorrell","year":"2019"},{"key":"2025120800370056400_ref029","doi-asserted-by":"publisher","first-page":"7","DOI":"10.18653\/v1\/d17-2002","article-title":"Argotario: computational argumentation meets serious games","author":"Habernal","year":"2017"},{"key":"2025120800370056400_ref030","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1016\/j.imavis.2019.06.001","article-title":"Detecting and classifying online dark visual propaganda","volume":"89","author":"Hashemi","year":"2019","journal-title":"Image and Vision Computing"},{"key":"2025120800370056400_ref031","first-page":"1","article-title":"Bots, #strongerin, and #brexit: computational propaganda during the UK-EU referendum","author":"Howard","year":"2016","journal-title":"arXiv preprint"},{"volume-title":"Junk News and Bots during the US Election: What Were Michigan Voters Sharing over Twitter","year":"2017","author":"Howard","key":"2025120800370056400_ref032"},{"key":"2025120800370056400_ref033","doi-asserted-by":"publisher","first-page":"244","DOI":"10.1609\/icwsm.v7i1.14443","article-title":"Dude, srsly?: the surprisingly formal nature of Twitter\u2019s language","author":"Hu","year":"2013"},{"key":"2025120800370056400_ref034","first-page":"1","article-title":"UniLeon-UniBO at IberLEF 2023 task DIPROMATS: RoBERTa-based models to climb up the propaganda tree in English and Spanish","author":"J\u00e1\u00f1ez-Martino","year":"2023"},{"key":"2025120800370056400_ref035","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1140\/epjds\/s13688-021-00308-4","article-title":"Characterizing partisan political narrative frameworks about COVID-19 on Twitter","volume":"10","author":"Jing","year":"2021","journal-title":"EPJ Data Science"},{"volume-title":"Propaganda and Persuasion","year":"2014","author":"Jowett","key":"2025120800370056400_ref036"},{"key":"2025120800370056400_ref037","first-page":"1415","article-title":"ApplicaAI at SemEval-2020 task 11: on RoBERTa-CRF, span CLS and whether self-training helps them","author":"Jurkiewicz","year":"2020"},{"key":"2025120800370056400_ref038","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2020.101075","article-title":"Multilingual stance detection in social media political debates","volume":"63","author":"Lai","year":"2020","journal-title":"Computer Speech and Language"},{"key":"2025120800370056400_ref039","first-page":"1","article-title":"RoBERTa: a robustly optimized BERT pretraining approach","author":"Liu","year":"2019","journal-title":"ArXiv preprint"},{"key":"2025120800370056400_ref040","unstructured":"Ma, E.\n           (2019), \u201cNLP augmentation\u201d, available at:\u00a0https:\/\/pypi.org\/project\/nlpaug\/0.0.5\/ (accessed\u00a01 September 2024)."},{"key":"2025120800370056400_ref041","doi-asserted-by":"crossref","first-page":"6064","DOI":"10.18653\/v1\/2024.findings-acl.363","article-title":"HQP: a human-annotated dataset for detecting online propaganda","author":"Maarouf","year":"2024","journal-title":"Findings of the Association for Computational Linguistics ACL 2024"},{"issue":"8","key":"2025120800370056400_ref042","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1371\/journal.pone.0221152","article-title":"Hate speech detection: challenges and solutions","volume":"14","author":"MacAvaney","year":"2019","journal-title":"PLoS One"},{"key":"2025120800370056400_ref043","doi-asserted-by":"crossref","first-page":"1","DOI":"10.7717\/peerj-cs.1248","article-title":"How to detect propaganda from social media? Exploitation of semantic and fine-tuned language models","volume":"9","author":"Malik","year":"2023","journal-title":"PeerJ Computer Science"},{"key":"2025120800370056400_ref044","first-page":"2758","article-title":"Sources of hallucination by large language models on inference tasks","author":"McKenna","year":"2023","journal-title":"Findings of the Association for Computational Linguistics: EMNLP 2023"},{"key":"2025120800370056400_ref045","doi-asserted-by":"publisher","first-page":"515","DOI":"10.1109\/iccke60553.2023.10326244","article-title":"BERT transformers multitask learning sarcasm and sentiment classification (BMSS)","author":"Molavi","year":"2023"},{"key":"2025120800370056400_ref046","first-page":"397","article-title":"Overview of DIPROMATS 2023: automatic detection and characterization of propaganda techniques in messages from diplomats and authorities of world powers","volume":"71","author":"Moral","year":"2023","journal-title":"Procesamiento del Lenguaje Natural"},{"key":"2025120800370056400_ref047","first-page":"9","article-title":"BERTweet: a pre-trained language model for English Tweets","author":"Nguyen","year":"2020"},{"key":"2025120800370056400_ref048","doi-asserted-by":"publisher","first-page":"155","DOI":"10.18653\/v1\/2023.bsnlp-1.18","article-title":"On experiments of detecting persuasion techniques in polish and Russian online news: preliminary study","author":"Nikolaidis","year":"2023"},{"key":"2025120800370056400_ref049","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"2025120800370056400_ref050","first-page":"2343","article-title":"SemEval-2023 task 3: detecting the category, the framing, and the persuasion techniques in online news in a multi-lingual setup","author":"Piskorski","year":"2023"},{"issue":"1-2","key":"2025120800370056400_ref051","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1007\/s12193-012-0102-z","article-title":"Comments by words, face and body","volume":"7","author":"Poggi","year":"2013","journal-title":"Journal on Multimodal User Interfaces"},{"key":"2025120800370056400_ref052","doi-asserted-by":"publisher","first-page":"433","DOI":"10.1145\/3677525.3678694","article-title":"MoralBERT: a fine-tuned language model for capturing moral values in social discussions","author":"Preniqi","year":"2024","journal-title":"Proceedings of the 2024 International Conference on Information Technology for Social Good"},{"key":"2025120800370056400_ref053","first-page":"1","article-title":"Investigating propaganda considering the discursive context of utterances","author":"Pritzkau","year":"2023"},{"key":"2025120800370056400_ref054","first-page":"2931","article-title":"Truth of varying shades: analyzing language in fake news and political fact-checking","author":"Rashkin","year":"2017"},{"key":"2025120800370056400_ref055","doi-asserted-by":"publisher","first-page":"16794","DOI":"10.18653\/v1\/2023.emnlp-main.1044","article-title":"Detecting propaganda techniques in code-switched social media text","author":"Salman","year":"2023","journal-title":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing"},{"key":"2025120800370056400_ref056","first-page":"1","article-title":"DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter","author":"Sanh","year":"2019","journal-title":"ArXiv preprint"},{"issue":"1","key":"2025120800370056400_ref057","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1007\/s13222-023-00437-2","article-title":"Automated multilingual detection of pro-kremlin propaganda in newspapers and Telegram posts","volume":"23","author":"Solopova","year":"2023","journal-title":"Datenbank-Spektrum"},{"key":"2025120800370056400_ref058","first-page":"1","article-title":"Large Language models for propaganda detection","author":"Sprenkamp","year":"2023","journal-title":"arXiv preprint"},{"key":"2025120800370056400_ref059","first-page":"1","article-title":"Efficient text-based propaganda detection via language model cascades","author":"Tian","year":"2023"},{"key":"2025120800370056400_ref060","doi-asserted-by":"publisher","first-page":"3433","DOI":"10.18653\/v1\/2022.naacl-main.251","article-title":"TWEETSPIN: fine-grained propaganda detection in social media using multi-view representations","author":"Vijayaraghavan","year":"2022"},{"issue":"6380","key":"2025120800370056400_ref061","doi-asserted-by":"publisher","first-page":"1146","DOI":"10.1126\/science.aap9559","article-title":"The spread of true and false news online","volume":"359","author":"Vosoughi","year":"2018","journal-title":"Science"},{"key":"2025120800370056400_ref062","first-page":"383","article-title":"What is propaganda, and what exactly is wrong with it","volume":"11","author":"Walton","year":"1997","journal-title":"Public Affairs Quarterly"},{"key":"2025120800370056400_ref063","first-page":"21","article-title":"Cross-domain learning for classifying propaganda in online contents","author":"Wang","year":"2020"},{"key":"2025120800370056400_ref064","first-page":"1","article-title":"Using GPT-2 to create synthetic data to improve the prediction performance of NLP machine learning classification models","author":"Whitfield","year":"2021","journal-title":"arXiv preprint"},{"key":"2025120800370056400_ref065","first-page":"38","article-title":"HuggingFace\u2019s transformers: state-of-the-art Natural Language processing","author":"Wolf","year":"2019"},{"key":"2025120800370056400_ref066","first-page":"1","article-title":"Experiments in detecting Persuasion techniques in the news","author":"Yu","year":"2019","journal-title":"Proceedings of the AI for Social Good Workshop at NeurIPS 2019"},{"key":"2025120800370056400_ref067","doi-asserted-by":"publisher","first-page":"1597","DOI":"10.26615\/978-954-452-072-4_179","article-title":"Interpretable propaganda detection in news articles","author":"Yu","year":"2021"},{"issue":"2","key":"2025120800370056400_ref068","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3161603","article-title":"Detection and resolution of rumours in social media: a survey","volume":"51","author":"Zubiaga","year":"2018","journal-title":"ACM Computing Surveys"}],"container-title":["Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/INTR-05-2024-0785\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/intr\/article-pdf\/35\/6\/2122\/10965139\/intr-05-2024-0785en.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/www.emerald.com\/intr\/article-pdf\/35\/6\/2122\/10965139\/intr-05-2024-0785en.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T05:37:10Z","timestamp":1765172230000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.emerald.com\/intr\/article\/35\/6\/2122\/1253108\/Together-we-can-do-it-A-roadmap-to-effectively"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,29]]},"references-count":68,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,12,9]]}},"URL":"https:\/\/doi.org\/10.1108\/intr-05-2024-0785","relation":{},"ISSN":["1066-2243","2054-5657"],"issn-type":[{"type":"print","value":"1066-2243"},{"type":"electronic","value":"2054-5657"}],"subject":[],"published":{"date-parts":[[2024,10,29]]}}}