http://dl.gi.de/handle/20.500.12116/36362 2026-07-21T14:09:11Z http://dl.gi.de/handle/20.500.12116/36405 Towards Understanding and Arguing with Classifiers: Recent Progress Shao, Xiaoting; Rienstra, Tjitze; Thimm, Matthias; Kersting, Kristian Machine learning and argumentation can potentially greatly benefit from each other. Combining deep classifiers with knowledge expressed in the form of rules and constraints allows one to leverage different forms of abstractions within argumentation mining. Argumentation for machine learning can yield argumentation-based learning methods where the machine and the user argue about the learned model with the common goal of providing results of maximum utility to the user. Unfortunately, both directions are currently rather challenging. For instance, combining deep neural models with logic typically only yields deterministic results, while combining probabilistic models with logic often results in intractable inference. Therefore, we review a novel deep but tractable model for conditional probability distributions that can harness the expressive power of universal function approximators such as neural networks while still maintaining a wide range of tractable inference routines. While this new model has shown appealing performance in classification tasks, humans cannot easily understand the reasons for its decision. Therefore, we also review our recent efforts on how to “argue” with deep models. On synthetic and real data we illustrate how “arguing” with a deep model about its explanations can actually help to revise the model, if it is right for the wrong reasons. 2020-01-01T00:00:00Z http://dl.gi.de/handle/20.500.12116/36399 Explaining Arguments with Background Knowledge Becker, Maria; Hulpuş, Ioana; Opitz, Juri; Paul, Debjit; Kobbe, Jonathan; Stuckenschmidt, Heiner; Frank, Anette Most information we consume as a society is obtained over the Web. News – often from questionable sources – are spread online, as are election campaigns; calls for (collective) action spread with unforeseen speed and intensity. All such actions have argumentation at their core, and the conveyed content is often strategically selected or rhetorically framed. The responsibility of critical analysis of arguments is thus tacitly transferred to the content consumer who is often not prepared for the task, nor aware of the responsibility. The ExpLAIN project aims at making the structure and reasoning of arguments explicit – not only for humans, but for Robust Argumentation Machines that are endowed with language understanding capacity. Our vision is a system that is able to deeply analyze argumentative text: that identifies arguments and counter-arguments, and reveals their internal structure, conveyed content and reasoning. A particular challenge for such a system is to uncover implicit knowledge which many arguments rely on. This requires human background knowledge and reasoning capacity, in order to explicate the complete reasoning of an argument. This article presents ongoing research of the ExpLAIN project that aims to make the vision of such a system a tangible aim. We introduce the problems and challenges we need to address, and present the progress we achieved until now by applying advanced natural language and knowledge processing methods. Our approach puts particular focus on leveraging available sources of structured and unstructured background knowledge, the automatic extension of such knowledge, the uncovering of implicit content, and reasoning techniques suitable for informal, everyday argumentation. 2020-01-01T00:00:00Z http://dl.gi.de/handle/20.500.12116/36402 Reconstructing Arguments from Noisy Text Dykes, Natalie; Evert, Stefan; Göttlinger, Merlin; Heinrich, Philipp; Schröder, Lutz Social media are of paramount importance to public discourse. RANT aims to contribute methods and formalisms for extracting, representing, and processing arguments from noisy text found in social media discussions, using a large corpus of pre-referendum Brexit tweets as a running case study. We identify recurring linguistic argumentation patterns in a corpus-linguistic analysis and formulate corresponding corpus queries to extract arguments automatically. Given the huge amount of social media data available, our approach aims at high precision at the possible price of low recall. Argumentation patterns are directly associated with logical patterns in a dedicated formalism and accordingly, individual arguments are directly parsed as logical formulae. The logical formalism for argument representation features a broad range of modalities capturing real-life modes of expression. We cast this formalism as a family of instance logics in the generic framework of coalgebraic logic and complement it by a flexible framework to represent relationships between arguments; including standard relations like attack and support but also relations extracted from metadata. Some relations are inferred from the logical content of individual arguments. We are in the process of developing suitable generalizations of various extension semantics for argumentation frameworks combined with corresponding algorithmic methods to allow for the automated retrieval of large-scale argumentative positions. 2020-01-01T00:00:00Z http://dl.gi.de/handle/20.500.12116/36404 News 2020-01-01T00:00:00Z