Student: Rodolpho da Silva Nascimento
Title: A Comparative Analysis Between Affective Lexicons in Portuguese
Advisor: Gustavo Paiva Guedes e Silva
Committee: Gustavo Paiva Guedes e Silva (president), Kele Teixeira Belloze (CEFET/RJ), Raimundo Santos Moura (UFPI)
Day/Time: April 28, 2021 / 10h.
Abstract: Affective Computing is a research area that studies the use of techniques that aim to enable a computer, making it capable of recognizing emotions through data analysis and interpretation. Sentiment Analysis (SA) is a relevant search field in this area, which corresponds to uses automatic tools capable of extracting subjective information in texts, such as opinions and feelings. There are works in SA that use Affective Lexicons (AF) to support automatic machine learning algorithms or as a main component. An AL is a set of words that capture the knowledge that speakers and listeners have about basic lexical expressions, forming a set of words that belong in human feelings context. However, few studies apply AL in the Portuguese language. This research aims to investigate available AL in Portuguese and evaluate their contributions in SA. In this regard, it is intended to provide resources for future work to develop new LA in Portuguese or improve existing ones. This research is also intended to be a reference for a better choice of AL in Portuguese, helping related works and contributing to better results in SA Portuguese tasks.