Defining textual signals to automatic classification of informative and opinionative texts on social networks

• Grant number: 23/12267-3
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: October 26, 2023
• End date: February 25, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Helena de Medeiros Caseli
• Grantee: Leticia Bossatto Marchezi
• Supervisor: Carolina Evaristo Scarton
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Institution abroad: University of Sheffield, England
• Associated to the scholarship: 22/16511-3 - Automatic classification of informative and opinionated texts on social networks, BP.IC

Abstract

In the contemporary digital era, the Internet's evolution has precipitated novel paradigms of communication, leading to the diversification of information dissemination channels. While the rise of social media platforms, notably Twitter, has democratized information sharing, a pertinent challenge lies in distinguishing personal opinionated content from factual, informative content. Despite its growing importance, studies focusing on the automatic differentiation of such content in the political context tweets, remain sparse. This project aims to elucidate the linguistic signals that are used to differentiate opinions from factual content, with a primary focus on Twitter posts in political domain. By harnessing such insights, this project endeavors to innovate heuristics and machine learning techniques rooted in linguistic principles, providing a new approach to weak label a large corpus composed of tweets. This work extends the Scientific Initiation project under development with FAPESP support (22/16511-3) and is also part of a bigger project (FAPESP's Research Grant 22/03090-0). (AU)