Design and Development of a Sentiment Analysis System on Facebook from Political Domain

Pablo Aramburu García. (2017). Design and Development of a Sentiment Analysis System on Facebook from Political Domain. Trabajo Fin de Titulación (TFG). ETSI Telecomunicación, Universidad Politécnica de Madrid.

Abstract:
This thesis is the result of a project whose main goal has been to develop and deploy a dashboard showing the information analysis’ results about political domain on Facebook, based on Web Components. To do so, the first task carried out has been the data extraction. This was developed on Python language, setting up a pipeline between Facebook, Senpy and Elasticsearch, managed by Luigi. This tool help us to analyse data and store it in Elasticsearch; making viable its access to view it. Right after the first task and using Sefarad, data has been deployed in widgets which show statistics estimated from it; helping users to read data quickly and easily on a striking interface. As a result, this project will allow us to make a Facebook research and analysis, making emphasis on emotions and sentiments that users show through it. This exploration will focus on the main Spanish political parties; but it can be applied to any social subject.