Mineração de dados aplicada a análise de desempenho de alunos no 5º ano do ensino fundamental

Abstract

Differences in the educational performance of students can be seen through evidence made byschools and evaluations of government educational systems, and this performance is directlylinked with socioeconomic of student and school structures as well as their teachers. One ofthe evaluations made by the INEP (National Institute of Research Studies) to evaluate theperformance of elementary school students, in which this work is dedicated, was created SAEB(System of Evaluation of Basic Education). Through the data provided by SAEB, a more detailedanalysis of student performance can be made. The aim of this work is to use computationaltechniques to analyze data from the SAEB referring to students of the 5th year of elementaryschool in the state public schools of pernambuco. In order to perform this work, the CRISP-DM(Cross Industry Standard Process for Data Mining) methodology was used, making use of all itssteps to better understand and organize the data. The work made use of the Weka software, whichin turn helped in the application of techniques and algorithms for the analysis of the database.The research was able to bring the development of a performance model to assist educationalmanagers and teachers in decision making.