Minería de texto de la web, de opinión pública y hechos referentes al barrio La Floresta

Abstract

The present project details the construction of a data analytics system oriented to text mining. The data analysis focuses on determining the topics of the texts, with the feelings and emotions expressed in them. This system is elaborated through a descriptive and predictive analysis approach. The data collected was gathered from web sources such as news web sites, blogs and social media with about the neighborhood La Floresta of Quito. The final product of the execution of the project is a system that has the functionality of classifying text according to the topic to which it belongs to, and identifying the feelings and emotions expressed in it. The system presents the results of the analysis graphically in a web interface. The results of the processing of data in the system show us that, according to the topics and the feelings of the texts, La Floresta is a united and organized neighborhood with a great cultural and artistic offer.