Times have changed, political discourse is no longer as it used to be, now data analysis
studies are gathered that inform the rulers of how to proceed before the crowd, and how
to address their public, through the analysis of feelings. It should not be forgotten that
those who vote for them are the citizens. In this study, the tweets of the different political
parties that make up the Spanish parliament are monitored, and an algorithm of sentiment
analysis is applied to them, in the context of a critical issue.
After the study of the State of the Art, this project has the goal of designing the Docker
system using an orchestrator such as Luigi and implementing a dashboard using Big Data
and Visualization technologies like Elasticsearch or Polymer Web W3C Components, which
show statistics regarding the activity of the political parties represented in the Spanish
parliament in the context of Twitter. These statistics are the activity within the network,
the number of daily tweets, the extracted topics spoken about, as well as a sentiment
analysis. To achieve this, we apply Sentiment Analysis with Natural Language Processing
These system characteristics allow us to time the input of data collected from Twitter,
get data related to Spanish politicians and analyze in more concrete ways, about the sentiments shown on Twitter about different terms in the COVID-19 emergency. In this study,
we try to focus on analyzing the political party from the citizen’s point of view. The dif-
ferent statistics collected by the system show an apparent ideological affinity for projecting
feelings on a topic. By breaking down the statistics, a chronological graphic carries out
on the activity and feelings shown in the tweets projected on a topic. We can check the
temporal evolution that the polarity about a concept may have had.
We show the guidelines, the activity, the feelings, and the position of a party on different
issues by applying the different technologies of Big Data analysis to the messages published
on the social network Twitter, all from a citizen’s point of view.