Dr. Edgar Casasola Murillo
Es estudiante:
No
Proyectos
II Coloquio Costarricense de Procesamiento del Lenguaje Natural
Número de proyecto:
745-B7-783
De 30/Jun/2017 hasta 29/Mar/2019
Análisis de sentimientos: una aproximación híbrida
Número de proyecto:
745-B4-048
De 06/Ene/2014 hasta 06/Ene/2017
Desarrollo de un modelo basado en extracción de características del texto para clasificar comentarios de opinión en lengua española según su polaridad
Número de proyecto:
326-B6-175
De 12/Ene/2016 hasta 31/Dic/2016
I Coloquio Costarricense de Procesamiento del Lenguaje Natural
Número de proyecto:
745-B5-754
De 01/Jul/2015 hasta 15/Dic/2016
Evaluación de algoritmos para análisis automático de sentimiento asociado a comentarios en español publicados en medios de opinión en línea
Número de proyecto:
326-B4-149
De 13/Ene/2014 hasta 27/Feb/2016
Desarrollo de una plataforma experimental para minería de opinión a partir de foros en línea
Número de proyecto:
326-B3-727
De 11/Mar/2013 hasta 03/Mar/2014
Construcción de una representación jerárquica de emociones: una taxonomía para el procesamiento automatizado del español
Número de proyecto:
745-C3-049
De 01/Ene/2023 hasta 31/Mar/2025
Publicaciones
Analysis of Semantic Shift Before and After COVID-19 in Spanish Diachronic Word Embeddings
Oct
2022
Descripción:
Words can shift their meaning across time. This case study shows the results obtained by the exploratory analysis of the semantic shifting on Spanish vocabulary using Diachronic Words Embeddings. Diachronic data consists of a 2018 Spanish corpus, before the COVID-19 outbreak, and a second corpus with documents from 2021. We focused on the semantic shift of three of the topics: COVID-19, masks and vaccines. This paper addresses the construction of the diachronic Spanish word embeddings model, as well as the results obtained by the analysis using a non-supervised distance vector technique. The results allowed to identify shifts related to increase in COVID-19 content.
Tipo de publicación: Conference Paper
Publicado en: 2022 XVLIII Latin American Computer Conference (CLEI)
Social Network Analysis for Automatic Ranking of Political Stakeholders: a Case Study
Nov
2022
Descripción:
This article exposes the way in which the creation of a new method for calculating the popularity of stake holders in social networks can support political data analysis experts. The definition of a new formula for estimating popularity allowed us to have a new method that, together with other previously existing ones, allows us to build a multidimensional interpretation of reality. The construction of a method that would seem like a computational scientific curiosity has significant impacts for experts who carry out political analysis. The new ranking algorithm called BOPRank made it possible to identify political actors in a different way than known algorithms. While a wellknown algorithm showed popularity as a result of the work of campaign teams on social networks, the new algorithm reflected popularity obtained as a result of the reaction of the public on social networks.
Tipo de publicación: Conference Paper
Publicado en: 2022 XVLIII Latin American Computer Conference (CLEI)