Semi-automatic approaches for exploiting shifter patterns in domain-specific sentiment analysis

• Tipo: Artigo em Revista Científica Internacional
• Data: 2022

Autores

Brazdil, Pavel (Autor) FEP; Muhammad, Shamsuddeen Hassan (Autor) Outra; Oliveira, Fátima (Autor) FLUP; Cordeiro, João (Autor) Outra; Silva, Maria de Fátima Henriques da (Autor) FLUP; Silvano, Maria da Purificação (Autor) FLUP; Leal, António (Autor) FLUP

Revista

Título: Mathematics

Vol. 10 Nº 18

Editora: MDPI

Indexação

ISI Web of Knowledge - 0 Citações

ISI Web of Science

Scopus - 0 Citações

Google Scholar

SHERPA/RoMEO

Worldcat OCLC

EBSCO

DOAJ

Current Contents

ProQuest

Classificação Científica

CORDIS: Humanidades > Ciências da linguagem > Linguística

Outras Informações

ID Authenticus: P-00X-4T0

DOI: 10.3390/math10183232

Abstract (EN): This paper describes two different approaches to sentiment analysis. The first is a form of symbolic approach that exploits a sentiment lexicon together with a set of shifter patterns and rules. The sentiment lexicon includes single words (unigrams) and is developed automatically by exploiting labeled examples. The shifter patterns include intensification, attenuation/downtoning and inversion/reversal and are developed manually. The second approach exploits a deep neural network, which uses a pre-trained language model. Both approaches were applied to texts on economics and finance domains from newspapers in European Portuguese. We show that the symbolic approach achieves virtually the same performance as the deep neural network. In addition, the symbolic approach provides understandable explanations, and the acquired knowledge can be communicated to others. We release the shifter patterns to motivate future research in this direction.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 24