3 documents found, page 1 of 1
Informed Data Selection Strategies for Few-Shot Learning on Imbalanced Data
Alcoforado, Alexandre; Okamura, Lucas; Ferraz, Thomas; Campos Fama, Israel; Dias Bueno, Bárbara; Veloso, Bruno Miguel; Reali Costa, Anna HelenaAcquiring high-quality annotated data remains one of the most significant challenges in Natural Language Processing (NLP), especially for supervised learning approaches. In scenarios where pre-existing labeled data is unavailable, common solutions like crowdsourcing and zero-shot approaches often fall short, suffering from limitations such as the need for large datasets and a lack of guarantees regarding annota...
ZeroBERTo: Leveraging Zero-Shot Text Classification by topic modeling
Alcoforado, Alexandre; Ferraz, Thomas Palmeira; Gerber, Rodrigo; Bustos, Enzo; Oliveira, André Seidel; Veloso, Bruno; Siqueira, Fabio LevyTraditional text classification approaches often require a good amount of labeled data, which is difficult to obtain, especially in restricted domains or less widespread languages. This lack of labeled data has led to the rise of low-resource methods, that assume low data availability in natural language processing. Among them, zero-shot learning stands out, which consists of learning a classifier without any p...
DEBACER: a method for slicing moderated debates
Ferraz, Thomas Palmeira; Alcoforado, Alexandre; Bustos, Enzo; Oliveira, André Seidel; Gerber, Rodrigo; Müller, Naíde; d’Almeida, André CorrêaSubjects change frequently in moderated debates with several participants, such as in parliamentary sessions, electoral debates, and trials. Partitioning a debate into blocks with the same subject is essential for understanding. Often a moderator is responsible for defining when a new block begins so that the task of automatically partitioning a moderated debate can focus solely on the moderator's behavior. In ...