Sentiment analysis for Twitter with Vader and TextBlob

Authors

DOI:

https://doi.org/10.35290/ro.v2n3.2021.494

Keywords:

Twitter, sentiments, algorithms, accuracy, python

Abstract

Sentiment analysis is a fundamental tool for the success of audience-oriented activities. Social networks have established themselves as a valid scenario to perform this analysis, especially Twitter, which offers a free API for data collection. The process for sentiment analysis includes download stages, using the Tweepy library, debugging by implementing methods to remove symbols that do not contribute to the sentiment of the tweet and analysis with two libraries: Vader and TextBlob. These return a percentage that defines whether the tweet is positive or negative; however, each one works with a different algorithm and training that causes discrepancy in the results, TextBlob presented greater accuracy. The final part of the analysis is the calculation of metrics: precision, accuracy, sensitivity, specificity and confusion matrix.

Downloads

Download data is not yet available.

References

Barrios Arce, J. (26 de Julio de 2019). La matriz de confusión y sus métricas. Health Big Data https://www.juanbarrios.com/la-matriz-de-confusion-y-sus-metricas/

Cambria, E., & Hussain, A. (2012). Sentic computing. marketing, 59(2), 557-577. https://doi.org/10.1007/978-94-007-5070-8_4

Hutto, C., & Gilbert, E. (2014). Vader: A parsimonious rule-based model for sentiment analysis of social media text. In Proceedings of the International AAAI Conference on Web and Social Media, 8(1).

Mejía Llano, J. (17 de marzo de 2021). ESTADÍSTICAS DE REDES SOCIALES 2021: USUARIOS DE FACEBOOK, INSTAGRAM, YOUTUBE, LINKEDIN, TWITTER, TIKTOK Y OTROS. https://www.juancmejia.com/marketing-digital/estadisticas-de-redes-sociales-usuarios-de-facebook-instagram-linkedin-twitter-whatsapp-y-otros-infografia/#4_Usuarios_activos_de_Twitter

Mendoza, M. (1994). Técnicas de observación directa para estudiar interacciones sociales infantiles entre los Toba. RUNA, archivo para las ciencias del hombre, 21(1), 241-262. https://doi.org/10.34096/runa.v21i1.1400

NLTK Project. (20 de abril de 2021). NLTK 3.6.2 documentation. https://www.nltk.org/

Pauli, P. A. (2019). Análisis de sentimiento - Comparación de algoritmos predictivos y métodos utilizando un lexicon expañol. [Tesis de grado, Instituto Tecnológico Buenos Aires] Repositorio ITBA. http://ri.itba.edu.ar/handle/123456789/1782

Roman, V. (25 de abril de 2019). Algoritmos Naive Bayes: Fundamentos e Implementación. Medium. https://medium.com/datos-y-ciencia/algoritmos-naive-bayes-fudamentos-e-implementaci%C3%B3n-4bcb24b307f

Romero Vega, R. (2021). Análisis de sentimientos en twitter para descubrir contenido xenófobo hacia los inmigrantes venezolanos en Ecuador. [Tesis de grado, Universidad Nacional de Loja] Repositorio UNL. https://dspace.unl.edu.ec/jspui/handle/123456789/23796

Twitter. (2021). Developer Platform. https://developer.twitter.com/en/docs/projects/overview

Universidad de Alcalá. (s.f). Lenguajes de programación para data science. https://www.master-data-scientist.com/lenguajes-programacion-data-science/

Vélez Bravo, (2020). Aplicativo para detectar conductas violentas en Twitter, a través de análisis de sentimientos. [Tesis de grado, Universidad Central del Ecuador]. Repositorio digital UCE. http://www.dspace.uce.edu.ec/handle/25000/20926

Published

2021-10-10

How to Cite

Alemán Viteri, S. B. (2021). Sentiment analysis for Twitter with Vader and TextBlob . ODIGOS JOURNAL, 2(3), 9–25. https://doi.org/10.35290/ro.v2n3.2021.494