Development of an affective load inference module for tweets in politics

• Grant number: 22/02472-6
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Effective date (Start): April 30, 2022
• Effective date (End): August 29, 2022
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Helena de Medeiros Caseli
• Grantee: Fernanda Malheiros Assi
• Supervisor: Carolina Evaristo Scarton
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Research place: University of Sheffield, England
• Associated to the scholarship: 21/07067-0 - Automatic inference of affective charge in social media posts, BP.IC

Abstract

Social media are increasingly becoming spaces where people share their opinions, emotions, and perceptions regarding countless topics, including politics, culture, and personal issues. In this context, several studies and research on natural language processing (NLP) have focused on social media text processing. Researches have, amongst others, studied the detection of emotions, learning emotion indicators from Twitter, emotion influence patterns, emotion classification and emotion lexicon creation. However, little effort has been made to analyze the affective load of political posts. Therefore, in this internship project, we propose creating a computational module capable of inferring the affective load of Bolsonaro's tweets and the replies he got during the COVID-19 pandemic. In order to achieve this, we will translate and expand the list of the 20 GEW emotion families into Portuguese. Using this translated and expanded version of GEW, our module will associate each tweet to a pair of coordinates on the GEW two-dimensional plane, indicating its positive/negative valence and power/control level. (AU)