Viviane Borelli 210 A similarity analysis was performed. It provided trees and graphs with associations between the words evoked and their relationships with one or more descriptive variables, known as metadata (Camargo and Justo, 2018). The tree (Figure 2) allows us to understand text construction structures and im- portant themes, as it shows the words that are close to and far from each other and their branches. Figure 2 - Similarity analysis (maximum tree) This type of analysis generates a graph of a network formed with the words in the text as nodes, and assigns weights to the edges that connect them (Romero and Borelli, 2021, p.7). In this case, we see the centrality of the term “message”, which is linked to “false”, “fake” and “circular”. These terms are connected
RkJQdWJsaXNoZXIy MjEzNzYz