Never Retreat, Never Retract: Argumentation Analysis for Political Speeches

Authors: Stefano Menini, Elena Cabrio, Sara Tonelli, Serena Villata

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We test and discuss our approach based on the analysis of documents issued by R. Nixon and J. F. Kennedy during 1960 presidential campaign. We rely on a supervised classifier to predict argument relations (i.e., support and attack), obtaining an accuracy of 0.72 on a dataset of 1,462 argument pairs.
Researcher Affiliation Academia Stefano Menini,1,3 Elena Cabrio,2 Sara Tonelli,1 Serena Villata2 1Fondazione Bruno Kessler, Trento, Italy 2 Universit e Cˆote d Azur, CNRS, Inria, I3S, France 3University of Trento, Italy
Pseudocode No The paper describes the system architecture and features but does not include any pseudocode or explicitly labeled algorithm blocks.
Open Source Code No The paper mentions that 'The script and the argumentation graphs... are available at https://dh.fbk.eu/resources/political-argumentation' but this refers to a script for converting data and the output graphs, not the full source code for the SVM model or feature extraction methodology.
Open Datasets Yes The dataset is available at https://dh.fbk.eu/resources/political-argumentation
Dataset Splits Yes We test the performance of the classification pipeline using the 1,462 manually annotated pairs with 10-fold cross-validation.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, or cloud instances) used for running the experiments.
Software Dependencies No The paper mentions software tools like 'Stanford Core NLP suite', 'LIBSVM', 'word2vec', and 'Excitement Open Platform', but does not provide specific version numbers for any of them.
Experiment Setup No The paper describes the features used and the choice of SVM with a radial kernel, but it does not specify concrete hyperparameter values (e.g., C, gamma for SVM) or other system-level training settings needed for full reproducibility.