ArgumenText
(Funding Period: 2017 - 2020)

Decision Support by Means of Automatically Extracting Natural Language Arguments from Big Data

Motivation

In order to make informed decisions, appropriate arguments are needed. However, the mere amount of information and the complexity of many questions frequently prevents us from finding all arguments that are relevant for a reasonable decision. Within the “Decision support by means of automatically extracting natural language arguments from big data” (in short, ArgumenText) project, the UKP Lab develops novel Argument Mining methods for extracting arguments from large and heterogeneous text sources in order to facilitate decision making processes.

In response to a user-defined search query, neural networks determine relevant arguments in realtime and summarize them in a comprehensive way. In contrast to conventional systems, an argumentative information system can show the reasons for or against a decision.

Goals

The key features of our envisioned argument mining system include:

  • Automatic extraction of arguments from large amounts of heterogeneous texts
  • Detecting pro and con arguments
  • Real-time summarization in natural language
  • Language adaptation of argument mining to German

In addition to supporting decision-making, our argumentative information system will provide an overview of current topics. Its real-time summary of natural language arguments simplifies a wide range of knowledge generation processes in areas such as journalism, politics, and economics.

Methods

In this project, we aim to validate argument mining methods for heterogeneous and large scale texts. Starting from a previously developed joint-modeling method for identifying argument structures in student essays (Stab and Gurevych 2017), we will develop robust end-to-end approaches for mining arguments from web-scale corpora. We also seek to extent our current methods to German using Language Adaptation and evaluate their robustness across various topics and text types. The resulting methods and software are intended to extract arguments from dynamic text sources in real-time and to prepare them for the user in a comprehensive summary.

Team

  • Prof. Dr. Iryna Gurevych, Principal Investigator
  • Dr. Christian Stab, project co-leader
  • Dr. Johannes Daxenberger, project co-leader
  • Dr. Steffen Eger
  • Dr. Tristan Miller
  • Chris Stahlhut
  • Christopher Tauchmann
  • Benjamin Schiller

Funding

The project is funded by the German Federal Ministry of Education and Research (BMBF) as part of the VIP+ program (“Validierung des technologischen und gesellschaftlichen Innovationspotenzials wissenschaftlicher Forschung”).

Publications

Schiller, Benjamin ; Daxenberger, Johannes ; Gurevych, Iryna (2021):
Stance Detection Benchmark: How Robust Is Your Stance Detection?
In: KI - Künstliche Intelligenz, Springer, ISSN 0933-1875,
DOI: 10.1007/s13218-021-00714-w,
[Article]

Schiller, Benjamin ; Daxenberger, Johannes ; Gurevych, Iryna (2021):
Aspect-Controlled Neural Argument Generation.
In: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 380-396,
ACL, 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics, virtual Conference, 06.-11.06.2021, ISBN 978-1-954085-46-6,
DOI: 10.18653/v1/2021.naacl-main.34,
[Conference or Workshop Item]

Thakur, Nandan ; Reimers, Nils ; Daxenberger, Johannes ; Gurevych, Iryna (2021):
Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks.
In: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 296-310,
ACL, 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics, virtual Conference, 06.-11.06.2021, ISBN 978-1-954085-46-6,
[Conference or Workshop Item]

Daxenberger, Johannes ; Gurevych, Iryna (2020):
Arguments as Social Good: Good Arguments in Times of Crisis.
AI for Social Good - AAAI Fall Symposium 2020, virtual Conference, 13.-14.11.2020, [Conference or Workshop Item]

Eger, Steffen ; Daxenberger, Johannes ; Gurevych, Iryna (2020):
How to Probe Sentence Embeddings in Low-Resource Languages: On Structural Design Choices for Probing Task Evaluation.
In: Proceedings of the 24th Conference on Computational Natural Language Learning, pp. 108-118,
ACL, 24th Conference on Computational Natural Language Learning (CoNLL 2020), virtual Conference, 19.-20.11., ISBN 978-1-952148-63-7,
[Conference or Workshop Item]

Daxenberger, Johannes ; Schiller, Benjamin ; Stahlhut, Chris ; Kaiser, Erik ; Gurevych, Iryna (2020):
ArgumenText: Argument Classification and Clustering in a Generalized Search Scenario.
In: Datenbank-Spektrum, 20 (2), pp. 115-121. Springer, ISSN 1618-2162,
DOI: 10.1007/s13222-020-00347-7,
[Article]

Rach, Niklas ; Matsuda, Yuki ; Daxenberger, Johannes ; Ultes, Stefan ; Yasumoto, Keiichi ; Minker, Wolfgang (2020):
Evaluation of Argument Search Approaches in the Context of Argumentative Dialogue Systems.
In: Proceedings of the 12th International Conference on Language Resources and Evaluation (LREC), pp. 513-522,
European Language Resources Association, Marseille, France, May 11-16 , 2020, [Conference or Workshop Item]

Simpson, Edwin ; Pfeiffer, Jonas ; Gurevych, Iryna (2020):
Low Resource Sequence Tagging with Weak Labels.
In: Proceedings of the AAAI Conference on Artificial Intelligence, pp. 8862-8869,
AAAI Press, 34th AAAI Conference on Artificial Intelligence (AAAI 2020), New York, USA, 07.-12.02.2020, e-ISSN 2374-3468,
DOI: 10.1609/aaai.v34i05.6415,
[Conference or Workshop Item]

Trautmann, Dietrich ; Daxenberger, Johannes ; Stab, Christian ; Schütze, Hinrich ; Gurevych, Iryna (2020):
Fine-Grained Argument Unit Recognition and Classification.
The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2020), New York, USA, 07.-12.02., ISBN 978-1-57735-835-0,
DOI: 10.1609/aaai.v34i05.6438,
[Conference or Workshop Item]

Reimers, Nils ; Gurevych, Iryna (2019):
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks.
pp. 3973-3983, The 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP 2019), Hong Kong, China, 03.12.2019-07.12.2019, [Conference or Workshop Item]

Rücklé, Andreas ; Moosavi, Nafise Sadat ; Gurevych, Iryna (2019):
Neural Duplicate Question Detection without Labeled Training Data.
pp. 1607-1617, The 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP 2019), Hong Kong, China, 03.11.2019--07.11.2019, [Conference or Workshop Item]

Stab, Christian ; Miller, Tristan ; Schiller, Benjamin ; Rai, Pranav ; Gurevych, Iryna (2018):
Cross-topic Argument Mining from Heterogeneous Sources.
Long Papers, In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pp. 3664-3674,
The 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, 31.10.2018--04.11.2018, [Conference or Workshop Item]

Eger, Steffen ; Rücklé, Andreas ; Gurevych, Iryna (2018):
PD3: Better Low-Resource Cross-Lingual Transfer By Combining Direct Transfer and Annotation Projection.
In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing,
5th Workshop on Argument Mining, Brussels, Belgium, 31.10.2018, [Conference or Workshop Item]

Rocha, Gil ; Stab, Christian ; Cardoso, Henrique Lopes ; Gurevych, Iryna (2018):
Cross-Lingual Argumentative Relation Identification: from English to Portuguese.
In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing,
5th Workshop on Argument Mining, Brussels, Belgium, 31.10.2018, [Conference or Workshop Item]

Stab, Christian ; Daxenberger, Johannes ; Stahlhut, Chris ; Miller, Tristan ; Schiller, Benjamin ; Tauchmann, Christopher ; Eger, Steffen ; Gurevych, Iryna (2018):
ArgumenText: Searching for Arguments in Heterogeneous Sources.
In: Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: System Demonstrations, pp. 21-25,
The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, New Orleans, Louisiana, 01.06.2018--06.06.2018, [Conference or Workshop Item]

Eger, Steffen ; Daxenberger, Johannes ; Stab, Christian ; Gurevych, Iryna (2018):
Cross-lingual Argumentation Mining: Machine Translation (and a bit of Projection) is All You Need!
In: Proceedings of the 27th International Conference on Computational Linguistics (COLING 2018), pp. 831-844,
The 27th International Conference on Computational Linguistics (COLING 2018), Santa Fe, USA, 20.08.2018--26.08.2018, [Conference or Workshop Item]

Schulz, Claudia ; Eger, Steffen ; Daxenberger, Johannes ; Kahse, Tobias ; Gurevych, Iryna (2018):
Multi-Task Learning for Argumentation Mining in Low-Resource Settings.
In: Proceedings of the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 35-41,
Association for Computational Linguistics, New Orleans, USA, [Conference or Workshop Item]

Stahlhut, Chris ; Stab, Christian ; Gurevych, Iryna (2018):
Pilot Experiments of Hypothesis Validation Through Evidence Detection for Historians.
In: CEUR Workshop Proceedings, 2167, In: Proceedings of the First Biennial Conference on Design of Experimental Search & Information Retrieval Systems, pp. 83-89,
First Biennial Conference on Design of Experimental Search & Information Retrieval Systems, Bertinoro, Italy, 28.08.2018--31.08.2018, [Conference or Workshop Item]

Daxenberger, Johannes ; Eger, Steffen ; Habernal, Ivan ; Stab, Christian ; Gurevych, Iryna (2017):
What is the Essence of a Claim? Cross-Domain Claim Identification.
In: Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 2055-2066,
Copenhagen, Denmark, [Conference or Workshop Item]

Lee, Ji-Ung ; Eger, Steffen ; Daxenberger, Johannes ; Gurevych, Iryna (2017):
UKP TU-DA at GermEval 2017: Deep Learning for Aspect Based Sentiment Detection.
In: Proceedings of the GermEval 2017 – Shared Task on Aspect-based Sentiment in Social Media Customer Feedback, pp. 22-29,
Berlin, Germany, [Conference or Workshop Item]

Eger, Steffen ; Daxenberger, Johannes ; Gurevych, Iryna (2017):
Neural End-to-End Learning for Computational Argumentation Mining.
Volume 1: Long Papers, In: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017), pp. 11-22,
Association for Computational Linguistics, Vancouver, Canada, [Conference or Workshop Item]