Temporal modeling of books' emotions using cognitive networks

• Grant number: 22/03683-0
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: June 25, 2022
• End date: September 24, 2022
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Diego Raphael Amancio
• Grantee: Giovana Daniele da Silva
• Supervisor: Giulio Rossetti
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Institute of information science and technologies "Alessandro Faedo" (ISTI), Italy
• Associated to the scholarship: 21/01744-0 - Using complex networks to characterize and identify the success of literary works, BP.IC

Abstract

As the amount of data available online increases every day, the need to access and process such data becomes quite significant. Because massive amounts of data cannot be - or are hardly - analyzed by humans, NLP tasks stand out as a way of programming computers to automatically process and analyze large amounts of data from natural language. However, although there are several methods that aim to analyze short texts (e.g. tweets or comments on online forums), few are those aimed at analyzing the longer ones (such as books, biographies, series or movie scripts). In this sense, the present research project proposes to detect similarities between books and patterns of success or failure through the temporal modeling of emotions of texts narratives using cognitive networks, thus providing a richer method of sentiment analysis throughout long texts. (AU)