Técnica para mineração de textos na análise de sentimentos: um estudo de caso em uma instituição de ensino técnico, tecnológico e superior

Abstract

This work proposes the application of data mining techniques in texts extracted from the social network Twitter for analysis of feelings. Companies have benefited from this tool with the generation of information about their products and / or services, as well as observing the behavior of users that can assist them in targeting promotions, launches, marketing, etc. With the increasing demand for the use of social networks as well as the use of unstructured texts, the interpretation of user evaluations by companies or stakeholders has become a difficult task. In this work, from data extracted and pre-processed from a social network, we intend to apply text mining approaches for data interpretation regarding the emotional level of users of a given educational institution. It is expected that the identification of the emotional state of individuals using this social network can help to identify the level of satisfaction of these individuals with regard to institution-related issues and also direct efforts / improvements to issues where emotions such as sadness, anger or disappointment in comments, and thereby remedy problems unknown or unnoticed by the educational institution.