RANT & RAND

Reconstructing Argument from Noisy Text & Newsworthy Debates

Reconstructing Arguments from Newsworthy Debates

(Third Party Funds Single)


Project leader:
Project members: , , ,
Start date: 2021-01-01
End date: 2023-12-31
Acronym: RAND
Funding source: DFG / Schwerpunktprogramm (SPP)
URL: https://www.linguistik.phil.fau.de/projects/rant/

Abstract:

Large portions of ongoing political debates are available in machine- readable form nowadays, ranging from the formal public sphere of parliamentary proceedings to the semi-public sphere of social media. This offers new opportunities for gaining a comprehensive overview of the arguments exchanged, using automated techniques to analyse text sources. The goal of the RANT/RAND project series within the priority programme RATIO (Robust Argumentation Machines) is to contribute to the automated extraction of arguments and argument structures from machine-readable texts via an approach that combines logical and corpus-linguistic methods and favours precision over recall, on the assumption that the sheer volume of available data will allow us to pinpoint prevalent arguments even under moderate recall. Specifically, we identify logical patterns corresponding to individual argument schemes taken from standard classifications, such as argument from expert opinion; essentially, these logical patterns are formulae with placeholders in dedicated modal logics. To each logical pattern we associate several linguistic patterns corresponding to different realisations of the formula in natural language; these patterns are developed and refined through corpus- linguistic studies and formalised in terms of corpus queries. Our approach thus integrates the development of automated argument extraction methods with work towards a better understanding of the linguistic aspects of everyday political argumentation. Research in the ongoing first project phase is focused on designing and evaluating patterns and queries for individual arguments, with a large corpus of English Twitter messages used as a running case study. In the second project phase, we plan to test the robustness of our approach by branching out into additional text types, in particular longer coherent texts such as newspaper articles and parliamentary debates, as well as by moving to German texts, which present additional challenges for the design of linguistic patterns (i.a. due to long- distance dependencies and limited availability of high-quality NLP tools). Crucially, we will also introduce similarity-based methods to enable complex reasoning on extracted arguments, representing the fillers in extracted formulae by specially tailored neural phrase embeddings. Moreover, we will extend the overall approach to allow for the high-precision extraction of argument structure, including explicit and implicit references to other arguments. We will combine these efforts with more specific investigations into the logical structure of arguments on how to achieve certain goals and into the interconnection between argumentation and interpersonal relationships, e.g. in ad-hominem arguments.

Publications:

Reconstructing Arguments from Noisy Text (DFG Priority Programme 1999: RATIO)

(Third Party Funds Single)


Project leader:
Project members: , , ,
Start date: 2018-01-01
End date: 2020-12-31
Acronym: RANT
Funding source: Deutsche Forschungsgemeinschaft (DFG)

Abstract:

Social media are of increasing importance in current public discourse. In RANT, we aim to contribute methods and formalisms for the extraction, representation, and processing of arguments from noisy text found in discussions on social media, using a large corpus of pre-referendum Twitter messages on Brexit as a running case study. We will conduct a corpus-linguistic study to identify recurring linguistic argumentation patterns and design corresponding corpus queries to extract arguments from the corpus, following a high-precision/low-recall approach. In fact, we expect to be able to associate argumentation patterns directly with logical patterns in a dedicated formalism and accordingly parse individual arguments directly as logical formulas. The logical formalism for argument representation will feature a broad range of modalities capturing real-life modes of expression such as uncertainty, agency, preference, sentiment, vagueness, and defaults. We will cast this formalism as a family of instance logics in the generic logical framework of coalgebraic logic, which provides uniform semantic, deductive and algorithmic methods for modalities beyond the standard relational setup; in particular, reasoning support for the logics in question will be based on further development of an existing generic coalgebraic reasoner. The argument representation formalism will be complemented by a flexible framework for the representation of relationships between arguments. These will include standard relations such as Dung's attack relation or a support relation but also relations extracted from metadata such as citation, hashtags, or direct address (via mention of user names), as well as relationships that are inferred from the logical content of individual arguments. The latter may take on a non-relational nature, involving, e.g., fuzzy truth values, preference orderings, or probabilities, and will thus fruitfully be modelled in the uniform framework of coalgebra that has already appeared above as the semantic foundation of coalgebraic logic. We will develop suitable generalizations of Dung's extension semantics for argumentation frameworks, thus capturing notions such as “coherent point of view” or “pervasive opinion”; in combination with corresponding algorithmic methods, these will allow for the automated extraction of large-scale argumentative positions from the corpus.

Publications: