Aplicação de análise de cluster em uma base de dados de escolas estaduais e municipais na região Sudeste

Abstract

The present work consisted in applying the fundamentals of cluster analysis through the combination of hierarchical and non-hierarchical algorithms. For this purpose, a database containing state and municipal public schools from the southeastern region of Brazil was used, in addition to the Python programming language for the practical part of the work. The objective of this work is to apply the principles and techniques of cluster analysis in an educational database, namely public schools of basic education in the Southeast region of Brazil, generating homogeneous groups of schools. It is concluded that it is possible to perform the segmentation of the data, with the generated groups having different characteristics among themselves. However, the hierarchical algorithm has a high computational cost and is not feasible for a large data set.