Seminar: Argumentation Mining (Winter 2025/2026)

Time & Location: Mi, 14:15-15:45 Gebäude C7 2 / Seminarraum -1.05


Teacher: Dr Volha Petukhova

***ANNOUNCEMENTS***

Registration in LSF, Veranstaltungsnummer 158609

Suitable for: Bachelor and Masters CoLi, LST and CS

Description

Enormous and ever growing digital content provides information where opinions, sentiment and arguments can be identified and analysed. News and social media content is searched to filter or weight the validity of statements, to identify the presence of fake news and false claims, to analyse opinions in public discussions, to detect opinion manipulation, to predict consumers sentiment, to study citizen engagement, and to recognize stance in political online debates. Arguments from legal, financial or medical documents are extracted to support professional decision-making. Natural argumentation is the focus of numerous educational scenarios assessing student’s essays and training argumentation and debate skills. The ability to construct good arguments and engage in argumentative discussions is assessed by argumentation systems focusing on training hypothetical reasoning, creating and structuring arguments, preventing opinion manipulation, detecting inconsistent arguments in online discussions and addressing different standpoints, attacking or sup porting claims with evidence as well as on the use of multimodal rhetorical devices.

Argumentation Mining (also known as “argument mining”) is an emerging research area within computational linguistics.  The students will learn fundamentals from argumentation theory and state-of-the-art methods from computational argumentation. In particular, we will focus on synergies between the fields of argument mining and natural language reasoning.

Organisation

Seminar meetings will take place once a week. After two weeks independent work each student will give a presentation on a selected topic (30 minutes + 10 minutes questions) – 4 credits. At the end of the course, to get 7 credits, some students has to prepare a short report (about 10 pages) and hand it for grading.

Grading: 40% based on the talk, 20% based on discussion participation and assignments; 40% based on the report

TOPICS

Identification, Assessment, and Analysis of Arguments: identification of argument components (e.g., premises and conclusions); structure analysis of arguments within and across documents; relation Identification between arguments and counterarguments (e.g., support and attack); creation and evaluation of argument annotation schemes, relationships to linguistic and discourse annotations, (semi-) automatic argument annotation methods and tools, and creation of argumentation corpora; assessment of arguments for various properties (e.g., stance, clarity)

Generation of Arguments, Multi-modal Argument Mining: automatic generation of arguments and their components; consideration of discourse goals in argument generation; argument mining and generation from multi-modal data

Mining and Analysis of different Genres and Domains of Arguments: argument mining in specific genres and domains (e.g., medicine, law, debate); modelling, assessing, and critically reflecting on the argumentative reasoning capabilities of Large Language Models

 Argumentation Techniques: for polarization, argumentative negotiation, neutralization

Interdisciplinary interfaces of Argument Mining: mining political discourse, by experts and laypeople; argument mining support for business communication; persuasion and convincingess from a psychological perspective; subjectivity, disagreements and perspectivism in argumentation

SELECTED PAPERS

  1. Peldszus, A., & Stede, M. (2013). From argument diagrams to argumentation mining in texts: A survey. International Journal of Cognitive Informatics and Natural Intelligence (IJCINI), 7(1), 1-31.
  2. Ye, Y., & Teufel, S. (2024, August). Computational Modelling of Undercuts in Real-world Arguments. In Proceedings of the 11th Workshop on Argument Mining (ArgMining 2024) (pp. 59-68).
  3. Falk, N., Jundi, I., Vecchi, E. M., & Lapesa, G. (2021, November). Predicting Moderation of Deliberative Arguments: Is Argument Quality the Key?. In Proceedings of the 8th Workshop on Argument Mining (pp. 133-141).
  4. Heinisch, P., Plenz, M., Opitz, J., Frank, A., & Cimiano, P. (2022, October). Data augmentation for improving the prediction of validity and novelty of argumentative conclusions. In Proceedings of the 9th Workshop on Argument Mining (pp. 19-33).
  5. Gemechu, D., Ruiz-Dolz, R., & Reed, C. (2024, August). Aries: A general benchmark for argument relation identification. In 11th Workshop on Argument Mining, ArgMining 2024 (pp. 1-14). Association for Computational Linguistics (ACL).
  6. Liu, Z., Guo, M., Dai, Y., & Litman, D. (2022). ImageArg: A multi-modal tweet dataset for image persuasiveness mining. arXiv preprint arXiv:2209.06416.
  7. Heinisch, P., Mindlin, D., & Cimiano, P. (2023, December). Unsupervised argument reframing with a counterfactual-based approach. In Proceedings of the 10th Workshop on Argument Mining (pp. 107-119).
  8. Korenčić, D., Chulvi, B., Casals, X. B., Toselli, A., Taulé, M., & Rosso, P. (2024). What distinguishes conspiracy from critical narratives? A computational analysis of oppositional discourse. Expert Systems, 41(11), e13671.
  9. Shahsavari, S., Holur, P., Wang, T., Tangherlini, T. R., & Roychowdhury, V. (2020). Conspiracy in the time of corona: automatic detection of emerging COVID-19 conspiracy theories in social media and the news. Journal of computational social science, 3(2), 279-317.
  10. Kaptein, M., & Van Helvoort, M. (2019). A model of neutralization techniques. Deviant behavior, 40(10), 1260-1285.
  11. Freedman, G., & Toni, F. (2024, August). Detecting Scientific Fraud Using Argument Mining. In Proceedings of the 11th Workshop on Argument Mining (ArgMining 2024) (pp. 15-28).
  12. Mezza, S., Wobcke, W., & Blair, A. (2024, August). Exploiting Dialogue Acts and Context to Identify Argumentative Relations in Online Debates. In Proceedings of the 11th Workshop on Argument Mining (ArgMining 2024) (pp. 36-45).
  13. Georgakopoulos, S. V., Tasoulis, S. K., Vrahatis, A. G., & Plagianakos, V. P. (2018, July). Convolutional neural networks for toxic comment classification. In Proceedings of the 10th hellenic conference on artificial intelligence (pp. 1-6).

SUGGESTED PAPERS

For any questions, please send an email to:
v.petukhova@lsv.uni-saarland.de
Use subject tag: [ArgMin_25/26]