Sentiment Analysis of News Articles in Spanish using Predicate Features
The automatic prediction of the course of action of agents involved in social or economic trends is an imperative challenge nowadays. However, it is a difficulttaskbecausestance or opinion is often spread throughout long, complex texts, such as news articles. The c...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Universidad del Valle
2019-07-01
|
Series: | Lenguaje |
Subjects: | |
Online Access: | http://revistalenguaje.univalle.edu.co/index.php/lenguaje/article/view/7937/11054 |
id |
doaj-07cbd0453d124aba920787e368f6b70e |
---|---|
record_format |
Article |
spelling |
doaj-07cbd0453d124aba920787e368f6b70e2020-11-25T03:31:54ZengUniversidad del ValleLenguaje2539-38042019-07-0147223526710.25100/lenguaje.v47i2.7937Sentiment Analysis of News Articles in Spanish using Predicate FeaturesAntonio Tamayo 0Julián Arias Londoño1Diego Burgos 2Gabriel Quiroz3Universidad de AntioquiaUniversidad de AntioquiaWake Forest UniversityUniversidad de AntioquiaThe automatic prediction of the course of action of agents involved in social or economic trends is an imperative challenge nowadays. However, it is a difficulttaskbecausestance or opinion is often spread throughout long, complex texts, such as news articles. The currentstudytests sentence predicates as features to automatically determine the writer’s stance in news articles. We capture the semantics and stance of the text by encoding features such as the attribute of copulative sentences, the predicate of transitive sentences, adjectival phrases, and the section of the article. Under the assumption that these features are informative enough to model the semantics of the text, each word sequence is disambiguated and assigned a sentiment value using weighting rules. Different experiments were run using either SentiWordNet and ML-Senticonto determine words’ sentiment. Feature vectors are automatically built to populate a database that is tested using two machine learning algorithms. An efficiency of 69% was achieved using a SVM with Gaussian kernel along with a feature selection strategy. This score outperformed the bag-of-words baseline in 12%. These results are promising considering that the sentiment analysis is performed on very complextexts written inSpanish.http://revistalenguaje.univalle.edu.co/index.php/lenguaje/article/view/7937/11054sentiment analysis;linguistic features; socio-politics;news articles; support vector machines; naïve bayes;dimension reduction. |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Antonio Tamayo Julián Arias Londoño Diego Burgos Gabriel Quiroz |
spellingShingle |
Antonio Tamayo Julián Arias Londoño Diego Burgos Gabriel Quiroz Sentiment Analysis of News Articles in Spanish using Predicate Features Lenguaje sentiment analysis;linguistic features; socio-politics;news articles; support vector machines; naïve bayes;dimension reduction. |
author_facet |
Antonio Tamayo Julián Arias Londoño Diego Burgos Gabriel Quiroz |
author_sort |
Antonio Tamayo |
title |
Sentiment Analysis of News Articles in Spanish using Predicate Features |
title_short |
Sentiment Analysis of News Articles in Spanish using Predicate Features |
title_full |
Sentiment Analysis of News Articles in Spanish using Predicate Features |
title_fullStr |
Sentiment Analysis of News Articles in Spanish using Predicate Features |
title_full_unstemmed |
Sentiment Analysis of News Articles in Spanish using Predicate Features |
title_sort |
sentiment analysis of news articles in spanish using predicate features |
publisher |
Universidad del Valle |
series |
Lenguaje |
issn |
2539-3804 |
publishDate |
2019-07-01 |
description |
The automatic prediction of the course of action of agents involved in social or economic trends is an imperative challenge nowadays. However, it is a difficulttaskbecausestance or opinion is often spread throughout long, complex texts, such as news articles. The currentstudytests sentence predicates as features to automatically determine the writer’s stance in news articles. We capture the semantics and stance of the text by encoding features such as the attribute of copulative sentences, the predicate of transitive sentences, adjectival phrases, and the section of the article. Under the assumption that these features are informative enough to model the semantics of the text, each word sequence is disambiguated and assigned a sentiment value using weighting rules. Different experiments were run using either SentiWordNet and ML-Senticonto determine words’ sentiment. Feature vectors are automatically built to populate a database that is tested using two machine learning algorithms. An efficiency of 69% was achieved using a SVM with Gaussian kernel along with a feature selection strategy. This score outperformed the bag-of-words baseline in 12%. These results are promising considering that the sentiment analysis is performed on very complextexts written inSpanish. |
topic |
sentiment analysis;linguistic features; socio-politics;news articles; support vector machines; naïve bayes;dimension reduction. |
url |
http://revistalenguaje.univalle.edu.co/index.php/lenguaje/article/view/7937/11054 |
work_keys_str_mv |
AT antoniotamayo sentimentanalysisofnewsarticlesinspanishusingpredicatefeatures AT julianariaslondono sentimentanalysisofnewsarticlesinspanishusingpredicatefeatures AT diegoburgos sentimentanalysisofnewsarticlesinspanishusingpredicatefeatures AT gabrielquiroz sentimentanalysisofnewsarticlesinspanishusingpredicatefeatures |
_version_ |
1724570960773775360 |