SIMILARITY ANALYSIS USED TO IDENTIFY RELATIONSHIPS BETWEEN WORDS WITHIN A TEXT CORPUS

Abstract

Similarity analysis is a technique used to identify and visualize the relationships between words or concepts within a textual corpus. Using co-occurrence algorithms, this methodology allows the construction of graphs that demonstrate the proximity and semantic interconnectedness between terms. This paper is part of the main author's PhD research in Public Health, which looked at theimpact of Covid-19 on the health of nursing professionals at a frontline hospital in Paraíba. The research, descriptive and exploratory in nature, used questionnaires with a qualitative approach and the IRaMuTeQ® software for data analysis. The results revealed that professionals, predominantly black and brown women, were the most affected by the pandemic, with high rates of mortality, stress, anxiety, depression and post-traumatic stress disorder. The data was obtained through a survey using a Google Forms® form about this context during the COVID-19 pandemic, given its serious and damaging consequences for the mental and physical health of those who provided direct care to infected patients.