Argael: Argument Annotation and Evaluation Tool

Segura-Tinoco, Andrés; Cantador, Iván

doi:10.2139/ssrn.4330222

Cited by 1 publication

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From this dataset, we selected the 40 most controversial proposals, and collaboratively searched for and annotated the arguments given by citizens in the proposals descriptions and comments, generating a first version of a corpus that we make publicly available 5 . To ease the manual argument annotation process, and store the identified arguments in a formal, structured format -including the components and relations of the arguments-, we used ARGAEL (Segura-Tinoco and Cantador, 2023), an easy-to-use, configurable desktop tool that we have developed to assist with the argument annotation and evaluation task, and which can be freely downloaded 6 .…”

Section: Case Studymentioning

confidence: 99%

Dimensionality Reduction for Machine Learning-based Argument Mining

Segura-Tinoco,

Cantador

2023

Proceedings of the 10th Workshop on Argument Mining

View full text Add to dashboard Cite

Recent approaches to argument mining have focused on training machine learning algorithms from annotated text corpora, utilizing as input high-dimensional linguistic feature vectors. Differently to previous work, in this paper, we preliminarily investigate the potential benefits of reducing the dimensionality of the input data. Through an empirical study, testing SVD, PCA and LDA techniques on a new argumentative corpus in Spanish for an underexplored domain (e-participation), and using a novel, rich argument model, we show positive results in terms of both computation efficiency and argumentative information extraction effectiveness, for the three major argument mining tasks: argumentative fragment detection, argument component classification, and argumentative relation recognition. On a space with dimension around 3-4% of the number of input features, the argument mining methods are able to reach 95-97% of the performance achieved by using the entire corpus, and even surpass it in some cases.

show abstract

Section: Case Studymentioning

confidence: 99%