Últimas noticias

Este martes 24 de junio se presentan los resultados del proyecto AMOR en Aranjuez en el curso de verano de la URJC organizado por CETINIA titulado "Human-centred Artificial Intelligence: How to bypass the Turing Tra ...

Hoy 12/12/2024 se presenta el proyecto AMOR en el UNICO I+D Project Meet-up Madrid organizado por el proyecto ELADAIS con la participación de los proyectos UNICO CLOUD financiados en la UPM (ELADAIS, MAP 6G, RISC ...

The article "To Click It or Not to Click It: An Italian Dataset for Neutralising Clickbait Headlines" has been presented at the Tenth Italian Conference on Computational Linguistics (CLiC-it 2024). The publication i ...

Canal GSI

El artículo A semantic similarity-based perspective of affect lexicons for sentiment analysis, por Oscar Araque, Ganggao Zhu y Carlos A. Iglesias has sido aceptado y publicado en la revista Knowledge-Based Systems. Esta revista se encuentra indexada en JCR: (Q1, 4.396). 

Referencia:

Oscar Araque, Ganggao Zhu, Carlos A. Iglesias, A semantic similarity-based perspective of affect lexicons for sentiment analysis, Knowledge-Based Systems, Volume 165, 2019, Pages 346-359, ISSN 0950-7051, https://doi.org/10.1016/j.knosys.2018.12.005(http://www.sciencedirect.com/science/article/pii/S0950705118305926).

Abstract: Lexical resources are widely popular in the field of Sentiment Analysis, as they represent a resource that directly encodes sentimental knowledge. Usually sentiment lexica are used for polarity estimation through the matching of words contained in a text and their associated lexicon sentiment polarities. Nevertheless, such resources have limitations in vocabulary coverage and domain adaptation. Besides, many recent techniques exploit the concept of distributed semantics, normally through word embeddings. In this work, a semantic similarity metric is computed between text words and lexica vocabulary. Using this metric, this paper proposes a sentiment classification model that uses the semantic similarity measure in combination with embedding representations. In order to assess the effectiveness of this model, we perform an extensive evaluation. Experiments show that the proposed method can improve Sentiment Analysis performance over a strong baseline, being this improvement statistically significant. Finally, some characteristics of the proposed technique are studied, showing that the selection of lexicon words has an effect in cross-dataset performance. 

Keywords:

Sentiment analysis; Sentiment lexicon; Semantic similarity; Word embeddings