Data mining using Damicore: case study of brazilian olympic athletes 'interviews

Abstract

The Olympic Studies Group, from the University of São Paulo, has been conducting research since the 2000s in the Olympic universe. One of them involves interviewing the Brazilian Olympic athletes. There is a hypothesis in the group that, through the use of computational tools, new subjects and knowledge may emerge from the analysis of these interviews. For this study, it was proposed to use the Damicore tool, precisely in the search for patterns that can indicate the paths that researchers can follow. Several tests were performed in order to determine an initial model of analysis, moving from a merely intuitive assumption to another, based on more concrete results. Patterns were detected in relation to the interviewers of some modalities, besides pointing out which modalities probably had relevant subjects, given their volume of interviews.