Análisis de sentimientos en comunidades digitales utilizando técnicas de big data para Determinar patrones de comportamiento orientado a fenómenos sociales

Abstract

ABSTRACT The prenent research work aims at analyzin the uners feeling of different digital communities, in order to extract behavior patterns of social phenomena that affect the community, determining the existence of revelant topics in this post-pandemic era, such as Covid-19 and politics. The analysis was carried out through the different phases of the CRISP-DM methodology, which were carried out in R, with the execution of the most outstanding libraries such as twitteR, rtweet, tidyverse, knitr each one of them fulfill different funtions. Data were collected from the online social networking Twiter. Next,an exploratory og the behavior data was conducted, in order to determine the percentage of positive, negative, and neutral tweets about each topic to be analyzed. Results show that text mining is a discipline that in its currents state can be useful for decision-making for both companies and individuals, nevertheless, it is susceptible to improvement in order to take advantage of the massive amount of opinios in text issued by users of the different social media sites. Keywords: CRISP-DM, feeling, digital communities, social phenomena