Aplicación para el análisis de tendencias en Twitter aplicando técnicas de PLN
Fecha
2022-07-15
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Jaén: Universidad de Jaén
Resumen
Este trabajo pone en valor la importancia de extraer conocimiento de la redes sociales, se ha realizado un
estudio comparativo de aplicaciones existentes en el mercado analizando sus pros y contras. La inclusión de
tecnologías de PLN para este tipo de sistemas suponen un factor diferencial en la calidad de sus resultados, por
ello se ha estudiado distintas técnicas y se ha implementado una aplicación que permite recuperar tweets, en
tiempo real o ya publicados con anterioridad, atendiendo a palabras clave, hashtags, localización y usuarios.
Con los resultados del análisis podemos conocer las palabras, usuarios y hashtags más influyentes, además de
conocer gracias al PLN el sentimiento, emociones y lenguaje de odio que pudieran expresar dichos tweets.
This work highlights the importance of extracting knowledge from social networks, a comparative study of existing applications in the market has been carried out, analyzing their pros and cons. The inclusion of NLP technologies for this type of system supposes a differential factor in the quality of its results, for this reason different techniques have been studied and an application has been implemented that allows retrieving tweets, in real time or previously published, attending to keywords, hashtags, location and users. With the results of the analysis we can know the most influential words, users and hashtags, in addition to knowing thanks to the NLP the sentiment, emotions and language of hate that these tweets may express.
This work highlights the importance of extracting knowledge from social networks, a comparative study of existing applications in the market has been carried out, analyzing their pros and cons. The inclusion of NLP technologies for this type of system supposes a differential factor in the quality of its results, for this reason different techniques have been studied and an application has been implemented that allows retrieving tweets, in real time or previously published, attending to keywords, hashtags, location and users. With the results of the analysis we can know the most influential words, users and hashtags, in addition to knowing thanks to the NLP the sentiment, emotions and language of hate that these tweets may express.