Understanding the power of social media: an investigation with opinion mining

Abstract

This work investigates how sentiment analysis in a simulated social network can identify users' opinions for various purposes, such as market research and brand monitoring. The goal is to test the effectiveness of automated analysis of large volumes of unstructured data. The research uses mixed methods and is descriptive, exploring the influence of sentiment patterns among users and reviewing data analysis on social networks and natural language processing. As a result, it was possible to have a richer understanding of frequently used words, as well as the associations between different themes and feelings in the simulated community.